Influence of gases on the integrity of packaging materials ...mediatum.ub.tum.de/doc/1353365/1353365.pdf · of the persons indicated they would spend $0.25 - $0.50 extra for an HPP

TECHNISCHE UNIVERSITÄT MÜNCHEN

Lehrstuhl für Lebensmittelverpackungstechnik

Influence of gases on the integrity of packaging materials

upon high pressure treatment

Dipl.-Ing. Univ.

Julia Maria Claudia Sterr

Vollständiger Abdruck der von der Fakultät Wissenschaftszentrum Weihenstephan für

Ernährung, Landnutzung und Umwelt der Technischen Universität München zur Erlangung des

akademischen Grades eines

Doktor-Ingenieurs (Dr.-Ing.)

genehmigten Dissertation.

Vorsitzender: Prof. Dr. rer. nat. Rudi F. Vogel

Prüfer der Dissertation: 1. Prof. Dr. rer. nat. H.-Chr. Langowski

2. Prof. Dr.-Ing. Jens-Peter Majschak

Diese Dissertation wurde am ___24.05.2017___ bei der Technischen Universität München

eingereicht und durch die Fakultät Wissenschaftszentrum Weihenstephan für Ernährung,

Landnutzung und Umwelt am ____20.09.2017___ angenommen.

II

DANKSAGUNG

III

Danksagung

Eine Promotion macht und schafft man nicht alleine. Ich habe das Glück, dass es so viele

Menschen in meinem Leben gibt, die an mich glauben und die mich auf meinem Weg

unterstützt haben. Vor allem, wenn es mir in manchen Momenten nicht schnell genug

voran ging, hatte ich immer wunderbare Menschen um mich, die mich mit

verschiedensten Methoden wieder motiviert und mir weitergeholfen haben.

Herrn Prof. Dr. rer. nat. H.-Ch. Langowski, meinem Doktorvater, möchte ich danken,

dass er mir die Möglichkeit gegeben hat an seinem Lehrstuhl zu promovieren und mir das

Thema überlassen hat. Seine gerechte und ausgeglichene Art waren mir ein großes

Vorbild. Dank seines umfassenden Wissens konnte er mir bei allen (un)denkbaren

fachlichen Fragen weiterhelfen. Auch für das entgegengebrachte Vertrauen und ein hohes

Maß an akademischer Freiheit möchte ich mich bedanken.

Meinem Zweitprüfer Prof. Dr.-Ing. Jens-Peter Majschak und dem Prüfungsvorsitzenden

Prof. Dr. rer. nat Rudi F. Vogel danke ich für die Übernahme des jeweiligen Amtes.

Bei meinen Kollegen möchte ich mich für die fröhliche, anregende und abwechslungs-

reiche Arbeitsatmosphäre bedanken. Dank euch ist die Zeit am Lehrstuhl nur so

verflogen. Ein besonderer Dank geht an Benedikt Fleckenstein, der mir mit seiner klugen,

besonnenen Art immer geholfen hat, wenn ich mal den roten Faden verloren habe.

Ohne die Kollegen aus der Abteilung Materialentwicklung des Fraunhofer Instituts für

Verfahrenstechnik und Verpackung hätte ich keinen meiner Versuche so durchführen

können. Vielen Dank, dass ihr immer Zeit für mich und meine wissenschaftlichen

Probleme hattet und mir mit Rat und Tat zur Seite gestanden seid. Danke auch für die

vielen extrudierten Folienkilometer.

Meiner Familie, besonders meiner Mutter Agnes und meiner Tante Frieda möchte ich für

die immerwährende Unterstützung in allen persönlichen und finanziellen Lebenslagen

danken. Danke, dass ihr mir immer euer Vertrauen geschenkt habt.

Meiner Simone möchte ich danken, dass sie mich, vor allem im letzten Abschnitt meiner

Arbeit, auf besondere Weise immer wieder motiviert hat. Danke, dass du für mich da bist.

IV

SCIENTIFIC CONTRIBUTIONS

V

Scientific contributions

Full Papers

The following peer reviewed publications were generated in the period of this work and

are related to the topic of the thesis (publications which are part of this thesis are indicated

in bold).

The doctoral candidate is the main author of three out of the four publications presented

in this thesis and shares in the fundamental and major part of conception, idea and

execution of all scientific experiments as well as the data analysis. The subsequent writing

of the manuscripts was exclusively her product. To the review paper about the effect of

high pressure processing on the integrity of polymeric packaging (Chapter 3) the doctoral

candidate contributed in form of a relevant co-authorship.

1. STERR, J.; FLECKENSTEIN, B. S.; LANGOWSKI, H.-C. (2017): The theory of

decompression failure in polymers during the high-pressure processing of

food. In Food Engineering Reviews. Published online: 13. October 2017.

DOI: 10.1007/s12393-017-9171-9

2. FLECKENSTEIN, B. S.; STERR, J.; LANGOWSKI, H.-C. (2014): The effect of

high pressure processing on the integrity of polymeric packaging - Analysis

and categorization of occurring defects. In Packaging Technology & Science

27 (2), pp. 83–103. DOI: 10.1002/pts.2018.

3. STERR, J.; FLECKENSTEIN, B. S.; LANGOWSKI, H.-C. (2015): The effect of

high pressure processing on tray packages with modified atmosphere. In

Food Engineering Reviews 7 (2), pp. 209–221. DOI: 10.1007/s12393-014-

9081-z.

4. STERR, J.; RÖTZER, K.; WECK, K.; WIRTH, A. L. K.; FLECKENSTEIN, B. S.;

LANGOWSKI, H.-C. (2015): In-situ measurement of oxygen concentration

under high pressure and the application to oxygen permeation through

polymer films. In Journal of Chemical Physics 143 (11), p. 114201. DOI:

10.1063/1.4931399.


VI

5. FLECKENSTEIN, B. S.; STERR, J.; LANGOWSKI, H.-C. (2016): The influence of

high pressure treatment and thermal pasteurization on the surface of polymeric

packaging films. In Packaging Technology & Science 29 (6), pp. 323–336. DOI:

10.1002/pts.2213.

6. RICHTER, T.; STERR, J.; JOST, V.; LANGOWSKI, H.-C. (2010): High pressure-

induced structural effects in plastic packaging. In High Pressure Research 30

(4), pp. 555–566. DOI: 10.1080/08957959.2010.531722.

7. STERR, J. (2013): Konfokale Raman-Spektroskopie - ein nützliches Tool auch

für die Getränkeindustrie. In Der Weihenstephaner 81 (3), pp. 127–129.

8. STERR, J.; FLECKENSTEIN, B. S. (2016): Hochdruckbehandelte Tray-

Verpackungen. Untersuchung von strukturellen Veränderungen mit Raman-

Spektroskopie. In Verpackungsrundschau 67 (4), pp. 72–73.

Further full papers and scientific contributions

1. GEBHARDT, R.; STEINHAUER, T.; MEYER, P.; STERR, J.; PERLICH, J.; KULOZIK,

U. (2012): Structural changes of deposited casein micelles induced by

membrane filtration. In Faraday Discussions 158, pp. 77–88.

2. STRIXNER, T.; STERR, J.; KULOZIK, U.; GEBHARDT, R. (2014): Structural study

on hen-egg yolk high density lipoprotein (hdl) granules. In Food Biophysics 9

(4), pp. 314–321.

3. ZHUANG, Y.; STERR, J.; KULOZIK, U.; GEBHARDT, R. (2015): Application of

confocal Raman microscopy to investigate casein micro-particles in blend

casein/pectin films. In International Journal of Biological Macromolecules 74,

pp. 44–48.

4. ZHUANG, Y.; STERR, J.; SCHULTE, A.; KULOZIK, U.; GEBHARDT, R. (2016):

Casein microparticles from blend films forming casein/α-tocopherol emulsion

droplets in solution. In Food Biophysics. DOI: 10.1007/s11483-016-9446-3.


VII

Oral presentations with first authorship

1. STERR, J. (2008): Verpackungen für hochdruckbehandelte Lebensmittel.

Anforderungen, Risiken, Chancen. Hochdrucktagung. Universität für

Bodenkultur Wien. Wien, Austria, 9/16/2008.

2. STERR, J. (2009): Characterization of crystallinity of polymers using Raman

Spectroscopy. Confocal Raman Imaging. Ulm, Germany, 9/29/2009.

3. STERR, J. (2012): The effect of high pressure on the morphology of polymers – a

Raman spectroscopic study. 7th International Conference on High Pressure

Bioscience and Biotechnology (HPBB2012). Otsu, 11/2/2012.

4. STERR, J. (2013): Raman spectroscopic study of high pressure induced structural

changes and defects in polymer films. European High Pressure Research Group

International Meeting (EHPRG 51). Queen Mary, University of London.

London, UK, 9/6/2013.

5. STERR, J. (2014): Possibilities and restrictions for the quantitative measurement

of Polyethylene with Raman Spectroscopy. ICORS International Conference on

Raman Spectroscopy. Jena, 8/10/2014.

6. STERR, J. (2014): Hochdruckbehandlung und MAP. Stellen Sie sich der

Herausforderung. Freisinger Tage - Fokus Fleisch und Wurstwaren – Produkte,

Verfahren und Verpackungen. Freising, Germany, 11/25/2014.

7. STERR, J. (2015): Hochdruckpasteurisation - Wie geht das? Einfluss auf Füllgüter

und Material. Barriere-Verbundfolien - Verbesserte Haltbarkeit von

Lebensmitteln. SKZ - ConSem GmbH. Würzburg, 9/23/2015.

Poster presentations with first authorship

1. STERR, J. (2013): The effect of high pressure processing on tray packages with

modified atmosphere. Nonthermal food processing workshop, Florianopolis,

Brazil, 09/30/2013.

VIII

CONTENT

IX

Content

Danksagung ..................................................................................................................... III

Scientific contributions ..................................................................................................... V

1 Introduction ................................................................................................................ 1

2 The theory of decompression failure in polymers during the high-pressure processing

of food ............................................................................................................................. 21

3 The effect of high pressure processing on the integrity of polymeric packaging –

Analysis and categorization of occurring defects ............................................................ 77

4 In-situ measurement of oxygen concentration under high pressure and the application

to oxygen permeation through polymer films ............................................................... 123

5 The effect of high pressure processing on tray packages with modified

atmosphere ..................................................................................................................... 145

6 Discussion and Conclusion .................................................................................... 173

7 Summary ................................................................................................................ 183

8 Zusammenfassung .................................................................................................. 187

Curriculum Vitae ........................................................................................................... 191

X

INTRODUCTION

1

1 Introduction

Throughout the history of mankind people have endeavoured to preserve foods in order

to survive winter cold, bad harvests and times of war. Methods such as chilling, drying,

salting, smoking and adding preserving agents have been used in the past and are still

used today to inhibit the growth of microorganisms. Microbial inactivation is nowadays

mostly achieved by heat treatment.

With industrialisation and the increased scientific knowledge in the area of health,

nutrition and technology, consumer demands for healthy, minimally processed products

have increased. Foods should contain minimal or if possible no synthetic preservatives

whilst simultaneously they should have enhanced shelf lives. In developed countries,

environmental sustainability as well as ethical criteria are also important for consumers

(PARDO, ZUFÍA 2012). For these reasons, innovative technologies for food preservation

have been developed in recent years. Besides treatment with for example radiation, pulsed

electric fields, pulsed ultraviolet light and non-thermal plasma, high pressure processing

(HPP) has become a significant low temperature method for food preservation and

microbial inactivation. According to a survey on upcoming trends in food preservation

and food processing methods, HPP is viewed as the most important innovative

commercial technology for the next ten years (JERMANN ET AL. 2015). More information

about alternative non-thermal preservation techniques can be found elsewhere (MORRIS

ET AL. 2007; PEREIRA, VICENTE 2010; ZHANG ET AL. 2011).

1.1 The high pressure processing of food

1.1.1 Fundamentals and applications

The principle of preservation using high pressure has been known since the late 19th

century (CERTES 1884; HITE 1899). Commercial applications, however, were only

developed much more recently when the cost and energy efficiency of equipment and

technology became acceptable. Indeed, the first pressurised foods became available 30

years ago in Japan (CHEFTEL 1995; FARKAS 2016) and have been available since 1993 in

Europe (KUROWSKA ET AL. 2016). More than 300 high pressure units are now installed in

industry worldwide (BROWN ET AL. 2016). In the United States, the estimated volume of

INTRODUCTION

2

sales now amounts to over $2 billion annually (FARKAS 2016). The current estimated

price per kilogram of an HPP product ranges from 0.06 – 0.60 US$ (BALASUBRAMANIAM

ET AL. 2016b; BROWN ET AL. 2016; MÚJICA-PAZ ET AL. 2011), which is assumed to be

about 0.06 – 0.20 US$ more than the costs arising from thermal processing

(BALASUBRAMANIAM, FARKAS 2008). A survey by HICKS ET AL. (2009) indicated that

consumers are willing to spend more money per item if the process makes the product

safer. After a brief explanation of the benefits of high pressure processing, the majority

of the persons indicated they would spend $0.25 - $0.50 extra for an HPP product. More

information about consumer acceptance of HPP products can be found elsewhere (BRUHN

2016).

Typical goods treated by high pressure treatment to prolong the shelf life are temperature-

sensitive foods such as fresh fruit products (smoothies, sauces etc.), cold meats (smoked

ham, sausages, etc.) and ready-to-eat meals. In addition, various other applications have

been developed in recent years. For example, it was found that the amount of salt in

sausages could be reduced and the texture and tenderness of meat products improved by

HPP. HPP is also used to open the shell of clams, mussels and oysters and to detach the

meat of crustaceans from the skeleton. Good summaries about the influence of pressure

on fruit, vegetables, dairy products, meat products and seafood have been given (JOFRÉ,

SERRA 2016; MATSER, TIMMERMANS 2016; TABILO-MUNIZAGA ET AL. 2016; TRUJILLO ET

AL. 2016). In general, HPP give food producers the opportunity to bring fresh and

minimally processed food products to the market which would not be possible with other

preservation techniques. The prolonged shelf life and improved sensory, structural and

optical qualities make HPP viable for both producers and for customers.

However, the legal framework concerning high pressure technology is still relatively

vague as there is no general coverage of food processed by high pressure in EU

regulations and there are still ongoing discussions amongst experts as to whether HPP

technology and HPP products are part of the Novel Foods Regulation 258/97 (NFR).

According to the NFR, a process is novel if it was not used before 15 May 1997 in a

significant quantity or if significant negative changes arise in the food (e.g. composition

or structure). Some experts are of the opinion that HPP should not be considered as a

novel technology because it was already used as a preservation technology before 1997.

However, due to the fact that the product itself and not the process must be approved by

INTRODUCTION

3

the regulation, every pressure treated food product should be verified to have no harmful

changes. In 1998 a high pressure treated fruit-based preparation of Danone was accepted

by the French Food Safety Agency and since that time HPP products have been sold by

companies in Europe referring to this decision and without any further authorization. An

extensive discussion about regulations for HPP has been given elsewhere (CHOLEWIŃSKA

2010; KUROWSKA ET AL. 2016).

Some of the most important advantages of non-thermal pressure treatment compared to

heat treatment are summarised below:

(1) No significant loss of nutritional or organoleptic value.

(2) The action of pressure is homogeneous and immediate.

(3) The effect is independent of the mass, shape and composition of the

product regardless of whether it is liquid or solid and regardless of whether

it comes with or without packaging.

(4) Compared to heating methods, the energy consumption is relatively small

(PEREIRA, VICENTE 2010; TEWARI 2007). Also, the global warming

potential and water depletion are smaller for HPP then for thermal

pasteurisation processes (PARDO, ZUFÍA 2012).

(5) The extended shelf life is similar to that obtained by thermal pasteurisation

(PARDO, ZUFÍA 2012).

(6) There are opportunities for the development of new products and

processes such as the detachment of shells from crustacean meat.

Further information about the high pressure processing of food can be found in the

literature (BALASUBRAMANIAM ET AL. 2016a; BALNY ET AL. 1992; HENDRICKX, KNORR

2002; JAY ET AL. 2005; RAY ET AL. 2001; TEWARI, JUNEJA 2007; ZHANG ET AL. 2011).

1.1.2 High pressure processing

Pressures of up to 8,000 bars are usually applied for industrial high pressure processes,

along with moderate or chilled temperatures. A typical pressurising process is a batch

process and is carried out as follows:

(1) Firstly pre-packed food is placed inside a cylindrical basket and put inside

the pressure vessel. The total loading capacity of industrial-scale systems

ranges from 35 to 525 litres (BALASUBRAMANIAM ET AL. 2016a).

INTRODUCTION

4

(2) The vessel is filled with pressure transmitting fluid, which in most cases

is water for industrial applications. Some pilot plants in laboratories use

glycerol-water mixtures or other low compressible liquids.

(3) Pressure build up is realised with pumps and intensifiers. Typical

pressurising rates are 1,000 to 3,000 bar/min, depending on factors such

as the number and power of pumps, degree of filling and packaging design,

especially if it concerns vacuum packaging or packaging with gaseous

headspace (MULTIVAC 10/16/2016; NGUYEN, BALASUBRAMANIAM

2011; SYED ET AL. 2012).

(4) The pressure holding time depends on the product and the desired

inactivation rates. Holding times from 2 to 30 min are commonplace.

(5) In most equipment decompression is realised in an uncontrolled way by

opening a valve. MULTIVAC HPP machinery utilises a patented method

for controlled pressure decrease in defined steps (“soft decompression”)

(RICHTER 2014).

(6) After reloading the basket, the packaged good is also dried in some

applications.

As high pressure equipment is a very cost-intensive investment, the processing time and

maximum pressure must be reduced to improve the return on investment. Here, batch

processing is one of the biggest disadvantages of the HPP industry due to the fact that the

maximum throughput is limited. Even though some semi-continuous and continuous

systems have been developed, these processes are only applicable for liquid goods and

the required energy is significantly higher than for a batch process (LELIEVELD,

HOOGLAND 2016). Other methods such as oscillating pressure treatment are available,

where several pressure cycles are realised consecutively. It has been found that this is

more effective for microorganism inactivation, but has poor time and cost efficiency for

industrial applications (JAY ET AL. 2005; PALOU ET AL. 1998).

The following sections explain the physical and chemical changes that are induced by

high pressure and the mechanisms of microorganism inactivation.

INTRODUCTION

5

1.1.3 Principle of high pressure processing

High pressure processing involves statically determined systems, namely isostatic

systems. According to the isostatic principle, “a force transported to the surface of a fluid

is equally transmitted through the contact surface” (MARTÍNEZ-MONTEAGUDO,

BALASUBRAMANIAM 2016). This means that all fluids are able to transport the pressure

homogeneously and almost instantaneously without friction. This also means that the

applied force, here the pressure, is transmitted to the packaging and the food in three

dimensions, independent of the shape, volume or structure (CHEFTEL 1992).

For further understanding of pressure effects on the products, their constituents and the

packaging, the principle of LE CHATELIER must be considered. This principle is a more

detailed explanation of the fact that a two-phase system tries to minimise the Gibbs energy

and therefore the point of equilibrium of the system exposed to pressure will be shifted

to the phase with the lower specific volume (NEVERS 2012). This is clear from Equation

1-1 which shows that the increase in Gibbs Energy G with increasing pressure is reduced

if the volume V decreases. Additionally, the principle of microscopic ordering states that

at constant temperature an increase in pressure increases the degree of ordering of

molecules of a given substance (HEREMANS 1992).

Equation 1-1

� = ��,�

This principle can explain some of the changes to molecules and structures under

pressure. The next section discusses the influence of pressure on chemical interactions.

1.1.4 Influence of pressure on molecules, bonds and chemical reactions

As already mentioned, ordering and volume reducing reactions are preferred under

pressure. The concomitant changes to interatomic distances can lead to destruction of

bonds if the bond energy is distance-dependent, due to distortion of the balance between

repulsive and attractive forces (MARTÍNEZ-MONTEAGUDO, BALASUBRAMANIAM 2016).

Based on the interatomic distances of some bonds and interactions, it was expected that

Coulomb and van der Waals interactions would be highly affected by pressure

(MARTÍNEZ-MONTEAGUDO, SALDAÑA 2014). It was also expected that hydrogen bonding,

solvation and hydrophobic interactions would be less affected and that there would be

essentially no influence on covalent bonding because this is almost independent of the

INTRODUCTION

6

distance in comparison to Coulomb or van der Waals interactions. Other workers

mentioned disruption to salt bridges and deprotonation of charged groups, which also

induces the ionisation of water and weak acids, resulting in a decrease in pH (CHEFTEL

1992; EARNSHAW ET AL. 1995; MIN, ZHANG 2007; TAUSCHER 1995).

All of these changes to intermolecular bonds can explain the conformational changes to

the secondary, tertiary and quaternary structures of proteins and other large molecules, in

some cases resulting in irreversible unfolding, denaturation, aggregation and

gelatinisation under pressure (TAUSCHER 1995). In general, most proteins are denatured

at pressures above 4,000 bars, but the denaturation depends on several other factors such

as the protein structure, pressure range, pH, temperature and additives, for example salt

and sugar (BALNY, MASSON 1993). Enzymes, which are technically deemed to be proteins

in most cases, can be activated as well as inactivated by pressure, depending on the

reaction volume. Also, the substrate activity can change (CHEFTEL 1992; TAUSCHER

1995). Changes to physical properties such as the melting point, density and viscosity are

further consequences of distortion of intermolecular distances (MARTÍNEZ-

MONTEAGUDO, SALDAÑA 2014).

In contrast to the marked effect of pressure on proteins, small molecules having covalent

bonds such as vitamins and polyphenols mostly stay unaffected. This advantage of HPP

is the reason for the minimal changes in flavour, colour and nutritional properties

compared to heat treatment. However, in recent studies it has become clear that the

stability of bioactive compounds is highly dependent on chemical reactions, and

especially on oxidation processes (MAHADEVAN, KARWE 2016). It has, however, already

been noted that there is a dearth of relevant scientific studies concerning the in-situ

measurement of chemical reactions under pressure (MARTÍNEZ-MONTEAGUDO, SALDAÑA

2014).

1.1.5 Influence of pressure on microorganisms

Vegetative forms of microorganisms are inactivated at pressures ranging from 3,000 to

5,000 bars, whereas spores may survive pressures exceeding 10,000 bars if the processing

temperature is not high enough (CHEFTEL 1995; GOULD 1996). A reason for this is the

relative low water content of spores and the resulting resistance to solvation and excessive

ionisation (EARNSHAW ET AL. 1995; KNORR 1995). It was found that pressure can induce

the germination of spores and a subsequent pressure treatment (“oscillating pressure

INTRODUCTION

7

treatment”) is able to destroy these germinated spores (KNORR 1995). Besides oscillating

pressure treatment, so-called pressure assisted thermal sterilisation (PATS) facilitates the

inactivation process at high pressures, especially for spores. Here, high temperatures

ranging from 90 to 120°C are employed. The PATS process will not be discussed further

in this thesis.

Besides depending on the type and initial numbers of microorganisms, the inhibitory

effect of HPP also depends on the physicochemical properties of the product and on the

process conditions such as the maximum pressure, temperature and pressure holding time.

Bacteria, for example, are more resistant to high pressures than yeast and mould (MIN,

ZHANG 2007). The inactivation effect of HPP can in general be improved by higher water

activity, by low acidity of the product and/or by adding preservatives (JOFRÉ, SERRA

2016).

In summary, several different and overlapping pressure effects exist for microbial

inactivation (EARNSHAW ET AL. 1995; KNORR 1995; MARTÍNEZ-MONTEAGUDO, SALDAÑA

2014; MIN, ZHANG 2007):

- Ionisation and concomitant precipitation of protein complexes

- ATPase inactivation or destabilisation (results in low internal pH)

- Physical modifications to membrane permeability and membrane protein function

and stability

- Collapse of intracellular vacuoles

- Separation of cell walls and cytoplasmic membranes

- Ribosomal destruction

- Mitochondrial damage

1.2 The role of packaging with high pressure processing

At every single stage, from harvest to consumption, spoilage of a product can occur. After

the food is processed, the packaging undertakes the task of protecting the product against

mechanical stress and physical, chemical and microbial contamination in order to

maximise the shelf life. Oxidation processes, for example, can be reduced by a good

active or passive gas barrier in the packaging film. Negative alterations due to light

absorption can be prevented by using opaque materials. More recently, active packaging

INTRODUCTION

8

has provided various additional functions such as humidity regulation and the absorption

of ethylene gas produced by fresh fruit. Besides its protective tasks, the packaging acts as

a medium for transferring information to the consumer about the ingredients, shelf life

and marketing matters. Often the design and appearance of packaging are driving factors

for consumer acceptance of a product and the decision to buy that product.

1.2.1 Combined effect of HPP and modified atmosphere packaging

The use of modified atmospheres in packaging technology is a standard method for

extending the shelf life and maintaining the quality of food products during storage.

Chemical and biochemical reactions are decelerated and so, depending on the gas and

food product, oxidation processes are inhibited, microbiological growth is slowed down

and colouring agents in fresh meats can be stabilized (MULLAN, MCDOWELL 2003). In

many cases, nitrogen also acts as an inexpensive gas for filling and support to prevent

collapse of the packaging. Nitrogen has a low permeability coefficient compared to other

gases such as CO2.

Modified atmosphere packaging (MAP) often includes CO2 for preservation. The gas acts

as a weak acid and dissolves in the food matrix, forming amongst other things bicarbonate

anions. A decreased internal pH in vegetative bacteria is a tentative explanation for the

observed inhibition of microorganism growth. However, the exact mechanism is still not

yet fully understood. It has, however, become clear that the growth of aerobic bacteria,

especially Gram-negative bacteria, is slowed down and that the inhibiting effect increases

at lower storage temperature (GOULD 1996; JAY ET AL. 2005; MULLAN, MCDOWELL

2003). Clostridia seem to be the most resistant to CO2. At high pressures it was found that

the antimicrobial effect of CO2 is improved by the additional extraction of intracellular

substances such as phospholipids, which is enhanced due to the increased solubility of

the gas in the membrane (EARNSHAW ET AL. 1995; KAMIHIRA ET AL. 1987). Regarding

HPP, the effect of CO2 on the inactivation of living cells was found to be dependent on

the kind of microorganisms, the water content, the state of CO2 and the addition of ethanol

or acetic acid as an entrainer to CO2 (KAMIHIRA ET AL. 1987).

It must be understood that the application of modified atmospheres, and especially CO2,

in combination with HPP is part of a hurdle concept, with synergistic interactions

improving the quality and safety of the pressurised products. MAP also provides benefits

from an environmental perspective because the energy requirements are very low (energy

INTRODUCTION

9

is only required for the packaging and gas injection process) compared to thermal

processing (PARDO, ZUFÍA 2012).

1.2.2 Influence of pressure on polymeric packaging

For high pressure processing, the food products are packed prior to treatment in flexible,

polymeric based pouches, trays or bottles. The ability to treat ready packed food is a

further advantage of HPP, avoiding contamination with pressure transmitting fluids and

costly aseptic filling. However, it must be ensured that the packaging still fulfils all

necessary requirements for food protection and has acceptable optical properties after

processing. Selection criteria for packaging materials are the barrier properties (O2 and

water vapour permeation), mechanical properties, structural characteristics (crystallinity,

density etc.), conformity with food contact regulations with respect to migration of

packaging constituents into the food and the overall integrity of the packaging (e.g.

delamination).

The focus here will be on the use of polymer based materials for HPP. Further information

about the use of paperboard as an HPP packaging material can be found elsewhere

(CANER ET AL. 2004; OCHIAI, NAKAGAWA 1992).

For a better understanding of all the effects which take place in packaging during and

after high pressure treatment a schematic classification (Figure 1-1) has been developed

(FLECKENSTEIN ET AL. 2014; RICHTER ET AL. 2010). These were differentiated into direct

and indirect effects. Direct effects are induced by the high pressure and indirect effects

are caused by compressed headspace gases or food.

Pressure induced damage includes reversible as well as irreversible structural changes to

the polymers. Delamination phenomena are due to large differences in the mechanical

properties of the materials. Indirect effects on the polymer packaging in the form of

crystallisation, plasticisation and macroscopic damage such as blistering and

delamination are induced by the rapid compression of gases accompanying high heating

rates and increased solubility of gases in the matrices. The increased solubility also

facilitates possible extraction of polymer components and the devolatilisation of volatile

food components. The solubility was also identified as being the main factor causing

decompression failure (“explosive decompression failure” XDF), accompanied by the

development of bubbles and blisters in oversaturated materials.

INTRODUCTION

10

Figure 1-1 Schematic classification of effects on polymeric packaging materials upon high pressure treatment (FLECKENSTEIN ET AL. 2014; RICHTER ET AL. 2010).

The influence of hydrostatic pressure on vacuum packaging has been extensively

reviewed (AYVAZ ET AL. 2016; FLECKENSTEIN ET AL. 2014; HAN 2007; JULIANO ET AL.

2010); see also Chapter 3. Only a short summary of the results will be given here.

Permeability: The oxygen permeability of common polymers was found to resist the

process in most cases. Some studies, however, found even improved barrier properties

upon pressurisation. Similar results were found for water vapour permeability. Solely

inorganic barrier layers showed weak pressure stability and/or film delamination due to

large differences in elasticity compared to the adjacent flexible polymers.

Mechanical properties: The mechanical properties were mostly found to be unaffected

or slightly improved after pressure treatment. Some samples showed increased tensile

strength, indicating higher rigidity. However, these results were without any practical

importance for industrial applications due to the low effect in absolute terms.

INTRODUCTION

11

Structural effects: Comparing the findings about pressure induced structural changes

from different research groups it was clear that the findings varied considerably from

group to group. In Chapter 3 these differences are put down to the differing physical and

chemical measurement principles that were used. Most of the methods (e.g. differential

scanning calorimetry DSC) provided only poor spatial resolution of crystallinity

measurements. Locally limited changes in crystallinity or density might not have been

detected by these methods. Additional information about changes to the surface and

surface structure of polymers can be found elsewhere (FLECKENSTEIN ET AL. 2016).

Migration: Only a few migration studies have been performed to investigate the total

migration of substances induced by high pressure. Some researchers have found there to

be little influence, with all values of total migration conforming to the requirements of

EU directive 90/128 for the global migration limit.

Regarding the scientific work on this topic, it can be concluded that common flexible

polymers are suitable as vacuum packaging for high pressure processing at moderate

temperatures. However, only a small amount of data is available about the suitability of

polymers as part of modified atmosphere packaging (MAP) for high pressure treatment.

Up until now, scientific information about the indirect effects induced by gases in a

supercritical state on polymeric packaging is still lacking. In 1992 authors started

mentioning physical damage such as delamination and/or visible changes in the materials

(bubbles/blisters) in the presence of headspace gases used for HPP (FRADIN ET AL. 1998;

MERTENS 1993; OCHIAI, NAKAGAWA 1992). Since then only a few studies have dealt with

possible failure mechanisms of modified atmosphere packaging at high pressures or tried

to identify the parameters responsible for packaging failure or shortcomings in the food

products.

INTRODUCTION

12

1.3 Motivation and objective

The use of modified atmosphere packaging for the high pressure processing of food

provides benefits in terms of improved product quality and shelf life. As this is already

used by industry, the suitability of the packaging and regulatory compliance should also

be verified. As already stated by some authors, there is still a lack of information about

the influence of headspace gases on the integrity of the packaging at high pressure.

The objective of this work was to provide the necessary information for safe use of

polymeric materials as part of modified atmosphere packaging for the high pressure

processing of food.

The first part of this work focuses on the mechanisms behind explosive decompression

failure. As mentioned previously, bubbles form in highly supersaturated materials on

decompression. However, the exact mechanisms are still not clear. A literature study

(Publication I, Chapter 2) was necessary to quantify and qualify the key parameters

responsible for the formation of bubbles and blisters. The transport properties of gases in

the polymers were identified as being the most important parameters, but it also became

clear that there is a lack of information about the permeation, solubility and diffusion of

gases in polymers during HPP and there were no in-situ measurement systems for high

hydrostatic pressure.

Accordingly, a new method for in-situ measurement of the oxygen concentration and

oxygen permeation through polymers at hydrostatic pressures up to 2,000 bars was

developed based on the principle of fluorescence quenching (Publication III, Chapter 4).

A reduction in oxygen permeability through a polyethylene film by a factor of between

35 and 70 at 2,000 bars applied pressure was put down to the reduction in free volume

and chain motion.

Besides the formation of bubbles, many other effects are possible when using modified

atmosphere packaging for HPP. A review article was prepared, classifying effects and

identifying missing but relevant criteria (Publication II, Chapter 3). Direct effects induced

by high pressure and indirect effects caused by the combined use of headspace gases and

pressure on monolayer films and multi-layered packaging were categorised (Figure 1-1).

Aspects such as sorption induced crystallisation and plasticisation, local thermal effects,

solution and extraction effects and decompression failure were discussed. In summary, it

INTRODUCTION

13

was concluded that there was a lack of information about irreversible structural changes

induced by supercritical gases. Studies on plastic constituent migration due to

supercritical CO2 could not be found in the literature. The influence of parameters such

as headspace volume, film thickness and depressurisation rate on the packaging integrity

and migration properties was also unclear (JULIANO ET AL. 2010).

Based on these summarized findings, experiments were undertaken to investigate

important factors such as the material properties, packaging design, process parameters

and migration processes (Publication IV, Chapter 5). The extraction potential of

headspace gases such as supercritical CO2 to the overall and specific migration of certain

substances under high pressure processing was also part of the study. Changes in the

oxygen permeability of organic and inorganic barrier layers were analysed after pressure

treatment. The influence of pressure on polymer structure in combination with and

without use of supercritical gases in monolayer and multilayer films and trays was

investigated using Raman spectroscopy. This non-destructive method is based on

detection of the inelastic scattering of monochromatic light due to molecular vibrations

and allows qualitative and quantitative analysis of structural changes. The method has

very high spatial resolution and allowed determination of the crystallinity and density of

different polymers in monolayer and multilayer packaging and films. The emphasis was

put on polyethylene due to the fact this polymer is widely used in the food industry and

for HPP applications, but was also the polymer experiencing most decompression failure.

Not only the crystalline content but also the amorphous content of polyethylene could be

distinguished and so crystallization as well as plasticization could be identified.

This thesis is structured as follows:

1. Critical review of the mechanisms of bubble formation based on explosive

decompression failure.

2. Critical review of damage mechanisms and pressure induced effects on polymeric

materials and packaging.

3. A new methodology for measuring oxygen permeation through polymers under

high hydrostatic pressure.

4. Measurement of the influence of headspace gases on the integrity and crystallinity

of polymers and the migration potential of supercritical CO2.

INTRODUCTION

14

1.4 Chapter bibliography

AYVAZ, H.; BALASUBRAMANIAM, V. M.; KOUTCHMA, T. (2016): High pressure effects

on packaging materials. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas,

Huub L. M. Lelieveld (Eds.): High pressure processing of food. Principles,

technology and applications. New York, Heidelberg, Dordrecht, London: Springer

(Food Engineering Series), pp. 73–93.

BALASUBRAMANIAM, V. M.; BARBOSA-CÁNOVAS, G. V.; LELIEVELD, H. L. M. (Eds.)

(2016a): High pressure processing of food. Principles, technology and applications.

New York, Heidelberg, Dordrecht, London: Springer (Food Engineering Series).

BALASUBRAMANIAM, V. M.; BARBOSA-CÁNOVAS, G. V.; LELIEVELD, H. L. M. (2016b):

High-pressure processing equipment for the food industry. In V. M.

Balasubramaniam, Gustavo V. Barbosa-Cánovas, Huub L. M. Lelieveld (Eds.): High

pressure processing of food. Principles, technology and applications. New York,

Heidelberg, Dordrecht, London: Springer (Food Engineering Series), pp. 39–66.

BALASUBRAMANIAM, V. M.; FARKAS, D. F. (2008): High-pressure food processing. In

Food Science and Technology International 14 (5), pp. 413–418. DOI:

10.1177/1082013208098812.

BALNY, C.; HAYASHI, R.; HEREMANS, K.; MASSON, P. (Eds.) (1992): High pressure and

biotechnology. Colloque Inserm. 224 volumes. Montrouge, France: John Libbey

EUROTEXT (European Seminar on High Pressure and Biotechnology).

BALNY, C.; MASSON, P. (1993): Effects of high pressure on proteins. In Food Reviews

International 9 (4), pp. 611–628. DOI: 10.1080/87559129309540980.

BROWN, P.; CARDONE, F.; MEYER, R.; POCCHIARI, M. (2016): High-pressure

inactivation of transmissible spongiform Encephalopathy agents (prions) in

processed meats. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas, Huub L.

M. Lelieveld (Eds.): High pressure processing of food. Principles, technology and

applications. New York, Heidelberg, Dordrecht, London: Springer (Food

Engineering Series), pp. 317–330.

BRUHN, C. M. (2016): Consumer acceptance of high-pressure processed products:

American perspective. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas,

INTRODUCTION

15




CANER, C.; HERNANDEZ, R. J.; PASCALL, M.; BALASUBRAMANIAM, V. M.; HARTE, B.

R. (2004): The effect of high-pressure food processing on the sorption behaviour of

selected packaging materials. In Packaging Technology & Science 17 (3), pp. 139–

153. DOI: 10.1002/pts.650.

CERTES, A. A. (1884): De l’action des hautes pressions sur les phenomènes de la

purefaction et sur Ia vitalite’ des microorganisms d’eau donce et d’eau de mer. In

Comptes Rendus 99, pp. 385–388.

CHEFTEL, J. C. (1992): Effects of high hydrostatic pressure on food constituents: an

overview. In High Pressure and Biotechnology 224, pp. 195–209.

CHEFTEL, J. C. (1995): Review: High-pressure, microbial inactivation and food

preservation. In Food Science and Technology International 1 (2-3), pp. 75–90.

CHOLEWIŃSKA, A. E. (2010): High pressure law. The legislation on high pressure

processing and other factors that may have an impact on HPP application in the EU

food industry. Wageningen University, Wageningen.

EARNSHAW, R. G.; APPLEYARD, J.; HURST, R. M. (1995): Understanding physical

inactivation processes. Combined preservation opportunities using heat, ultrasound

and pressure. In International Journal of Food Microbiology 28 (2), pp. 197–219.

DOI: 10.1016/0168-1605(95)00057-7.

FLECKENSTEIN, B. S.; STERR, J.; LANGOWSKI, H.-C. (2014): The effect of high pressure

processing on the integrity of polymeric packaging - Analysis and categorization of

occurring defects. In Packaging Technology & Science 27 (2), pp. 83–103. DOI:

10.1002/pts.2018.

FLECKENSTEIN, B. S.; STERR, J.; LANGOWSKI, H.-C. (2016): The influence of high

pressure treatment and thermal pasteurization on the surface of polymeric packaging

films. In Packaging Technology & Science 29 (6), pp. 323–336. DOI:

10.1002/pts.2213.

INTRODUCTION

16

FRADIN, J. F.; LE BAIL, A.; SANZ, P. D.; MOLINA-GARCÍ-A, A. D. (1998): Note.

Behaviour of packaging materials during high pressure thawing. In Food Science and

Technology International 4 (6), pp. 419–424.

GOULD, G. W. (1996): Methods for preservation and extension of shelf life. In

International Journal of Food Microbiology 33 (1), pp. 51–64. DOI: 10.1016/0168-

1605(96)01133-6.

HAN, J. H. (Ed.) (2007): Packaging for nonthermal processing of food. IFT Press. 1st.

Iowa: Blackwell Pub.; IFT Press.

HENDRICKX, M. E. G.; KNORR, D. (2002): Ultra high pressure treatments of foods. New

York: Kluwer Academic/Plenum Publishers (Food Engineering Series).

HEREMANS, K. (1992): From living systems to biomolecules. In Claude Balny,

Rikimaru Hayashi, Karel Heremans, Patrick Masson (Eds.): High pressure and

biotechnology. 224 volumes. Montrouge, France: John Libbey EUROTEXT

(European Seminar on High Pressure and Biotechnology), pp. 37–44.

HICKS, D. T.; PIVARNIK, L. F.; MCDERMOTT, R.; RICHARD, N.; HOOVER, D. G.; KNIEL,

K. E. (2009): Consumer awareness and willingness to pay for high-pressure

processing of ready-to-eat food. In Journal of Food Science Education 8 (2), pp. 32–

38. DOI: 10.1111/j.1541-4329.2009.00069.x.

HITE, B. H. (1899): The effects of pressure in the preservation of milk. In West Virginia

Agricultural Experimental Station Bulletin 58, pp. 15–35.

JAY, J. M.; GOLDEN, D. A.; LOESSNER, M. J. (2005): Modern Food Microbiology. 7. ed.

New York, NY: Springer (Food science text series).

JERMANN, C.; KOUTCHMA, T.; MARGAS, E.; LEADLEY, C.; ROS-POLSKI, V. (2015):

Mapping trends in novel and emerging food processing technologies around the

world. In Innovative Food Science & Emerging Technologies. DOI:

10.1016/j.ifset.2015.06.007.

JOFRÉ, A.; SERRA, X. (2016): Processing of meat products utilizing high pressure. In V.

M. Balasubramaniam, Gustavo V. Barbosa-Cánovas, Huub L. M. Lelieveld (Eds.):

High pressure processing of food. Principles, technology and applications. New

INTRODUCTION

17

York, Heidelberg, Dordrecht, London: Springer (Food Engineering Series), pp. 591–

624.

JULIANO, P.; KOUTCHMA, T.; SUI, Q.; BARBOSA-CÁNOVAS, G. V.; SADLER, G. (2010):

Polymeric-based food packaging for high-pressure processing. In Food Engineering

Reviews 2 (4), pp. 274–297.

KAMIHIRA, M.; TANIGUCHI, M.; KOBAYASHI, T. (1987): Sterilization of microorganisms

with supercritical carbon dioxide. In Agricultural and Biological Chemistry 51 (2),

pp. 407–412. DOI: 10.1080/00021369.1987.10868053.

KNORR, D. (1995): Hydrostatic pressure treatment of food: microbiology. In G. W.

Gould (Ed.): New methods of food preservation. 1st ed. London, New York: Blackie

Academic & Professional, pp. 159–175.

KUROWSKA, A.; SZAJKOWSKA, A.; VAN DER MEULEN, B. (2016): EU regulatory

approach to high-pressure processing. In V. M. Balasubramaniam, Gustavo V.

Barbosa-Cánovas, Huub L. M. Lelieveld (Eds.): High pressure processing of food.

Principles, technology and applications. New York, Heidelberg, Dordrecht, London:

Springer (Food Engineering Series), pp. 717–732.

LELIEVELD, H. L. M.; HOOGLAND, H. (2016): Continuous high-pressure processing to

extend product shelf life. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas,




MAHADEVAN, S.; KARWE, M. V. (2016): Effect of high-pressure processing on

bioactive compounds. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas,



(Food Engineering Series).

MARTÍNEZ-MONTEAGUDO, S. I.; BALASUBRAMANIAM, V. M. (2016): Fundamentals and

applications of high-pressure processing technology. In V. M. Balasubramaniam,

Gustavo V. Barbosa-Cánovas, Huub L. M. Lelieveld (Eds.): High pressure

processing of food. Principles, technology and applications. New York, Heidelberg,

Dordrecht, London: Springer (Food Engineering Series).

INTRODUCTION

18

MARTÍNEZ-MONTEAGUDO, S. I.; SALDAÑA, M. D. A. (2014): Chemical reactions in food

systems at high hydrostatic pressure. In Food Engineering Reviews 6 (4), pp. 105–

127.

MATSER, A.; TIMMERMANS, R. (2016): High-pressure effects on fruits and vegetables.

In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas, Huub L. M. Lelieveld

(Eds.): High pressure processing of food. Principles, technology and applications.

New York, Heidelberg, Dordrecht, London: Springer (Food Engineering Series),

pp. 541–552.

MERTENS, B. (1993): Packaging aspects of high-pressure food processing technology.

In Packaging Technology & Science 6 (1), pp. 31–36. DOI:

10.1002/pts.2770060106.

MIN, S.; ZHANG, Q. H. (2007): Packaging for high-pressure processing, irradiation, and

pulsed electric field processing. In Jung H. Han (Ed.): Packaging for nonthermal

processing of food. 1st. Iowa: Blackwell Pub.; IFT Press, pp. 67–86.

MORRIS, C.; BRODY, A. L.; WICKER, L. (2007): Non-thermal food

processing/preservation technologies: a review with packaging implications. In

Packaging Technology & Science 20 (4), pp. 275–286. DOI: 10.1002/pts.789.

MÚJICA-PAZ, H.; VALDEZ-FRAGOSO, A.; SAMSON, C. T.; WELTI-CHANES, J.; TORRES, J.

A. (2011): High-pressure processing technologies for the pasteurization and

sterilization of foods. In Food Bioprocess Technol 4 (6), pp. 969–985. DOI:

10.1007/s11947-011-0543-5.

MULLAN, M.; MCDOWELL, D. (2003): Modified atmosphere packaging. In Richard

Coles, Derek McDowell, Mark L. Kirwan (Eds.): Food packaging technology. Boca

Raton: Blackwell Publishing Ltd; CRC Press (Packaging technology series),

pp. 303–331.

MULTIVAC (2016): Pressurizing rate of MULTIVAC equipment. Personal

communication to Julia Sterr, 10/16/2016.

NEVERS, N. de (2012): Physical and chemical equilibrium for chemical engineers. 2. ed.

Hoboken, NJ: AICHe/Wiley.

INTRODUCTION

19

NGUYEN, L. T.; BALASUBRAMANIAM, V. M. (2011): Fundamentals of food processing

using high pressure. In Q. Howard Zhang, Gustavo V. Barbosa-Cánovas, V. M.

Balasubramaniam, C. Patrick Dunne, D. F. Farkas, Yuan, James T. C. (Eds.):

Nonthermal processing technologies for food. Chichester, West Sussex: Wiley-

Blackwell, pp. 3–19.

OCHIAI, S.; NAKAGAWA, Y. (1992): Packaging for high pressure food processing. In

Claude Balny, Rikimaru Hayashi, Karel Heremans, Patrick Masson (Eds.): High

pressure and biotechnology. 224 volumes. Montrouge, France: John Libbey

EUROTEXT (European Seminar on High Pressure and Biotechnology), pp. 515–

519.

PALOU, E.; LÓPEZ-MALO, A.; BARBOSA-CÁNOVAS, G. V.; WELTI-CHANES, J.;

SWANSON, B. G. (1998): Oscillatory high hydrostatic pressure inactivation of

Zygosaccharomyces baili. In Journal of Food Protection 61 (9), pp. 1213–1215.

PARDO, G.; ZUFÍA, J. (2012): Life cycle assessment of food-preservation technologies.

In Journal of Cleaner Production 28, pp. 198–207. DOI:

10.1016/j.jclepro.2011.10.016.

PEREIRA, R. N.; VICENTE, A. A. (2010): Environmental impact of novel thermal and

non-thermal technologies in food processing. In Food Research International 43 (7),

pp. 1936–1943. DOI: 10.1016/j.foodres.2009.09.013.

RAY, B.; KALCHAYANAND, N.; DUNNE, C. P.; SIKES, A. (2001): Microbial destruction

during hydrostatic pressure processing of food: IOS Press (338).

RICHTER, T. (2014): Device and method for the high-pressure treatment of products.

Applied for by MULTIVAC SEPP HAGGENMUELLER GMBH & CO. KG on

5/15/2014. Patent no. US8771773B2.

RICHTER, T.; STERR, J.; JOST, V.; LANGOWSKI, H.-C. (2010): High pressure-induced

structural effects in plastic packaging. In High Pressure Research 30 (4), pp. 555–

566. DOI: 10.1080/08957959.2010.531722.

SYED, Q.-A.; REINEKE, K.; SALDO, J.; BUFFA, M.; GUAMIS, B.; KNORR, D. (2012):

Effect of compression and decompression rates during high hydrostatic pressure

INTRODUCTION

20

processing on inactivation kinetics of bacterial spores at different temperatures. In

Food Control 25 (1), pp. 361–367. DOI: 10.1016/j.foodcont.2011.10.061.

TABILO-MUNIZAGA, G.; AUBOURG, S.; PÉREZ-WON, M. (2016): Pressure effects on

seafoods. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas, Huub L. M.

Lelieveld (Eds.): High pressure processing of food. Principles, technology and

applications. New York, Heidelberg, Dordrecht, London: Springer (Food

Engineering Series), 625-670.

TAUSCHER, B. (1995): Pasteurization of food by hydrostatic high pressure: Chemical

aspects. In Z Lebensm Unters Forsch 200 (1), pp. 3–13. DOI: 10.1007/BF01192901.

TEWARI, G. (2007): High-Pressure Processing of Foods. In Gaurav Tewari, Vijay K.

Juneja (Eds.): Advances in thermal and non-thermal food preservation. Iowa,

Oxford, Victoria: Blackwell Publishing Ltd, pp. 203–239.

TEWARI, G.; JUNEJA, V. K. (Eds.) (2007): Advances in thermal and non-thermal food

preservation. Iowa, Oxford, Victoria: Blackwell Publishing Ltd.

TRUJILLO, A. J.; FERRAGUT, V.; JUAN, B.; ROIG-SAGUÉS, A. X.; GUAMIS, B. (2016):

Processing of dairy products utilizing high pressure. In V. M. Balasubramaniam,

Gustavo V. Barbosa-Cánovas, Huub L. M. Lelieveld (Eds.): High pressure

processing of food. Principles, technology and applications. New York, Heidelberg,

Dordrecht, London: Springer (Food Engineering Series), pp. 553–590.

ZHANG, Q. H.; BARBOSA-CÁNOVAS, G. V.; BALASUBRAMANIAM, V. M.; DUNNE, C. P.;

FARKAS, D. F.; Yuan, James T. C. (Eds.) (2011): Nonthermal processing

technologies for food. IFT Press. Chichester, West Sussex: Wiley-Blackwell.

PUBLICATION I CHAPTER 2

21

2 The theory of decompression failure in polymers during the

high-pressure processing of food

This review article is dealing with the influence of gases on polymers (thermoplastics,

elastomers etc.) under very high pressures up to several thousands of bars and subsequent

decompression. This topic is an important and crucial part of several current application

and research areas like in petroleum industry (“explosive decompression failures” XDF),

in fuel cell vehicles in hydrogen infrastructure, the foaming of polymers, in high pressure

processing of packed food (HPP) or even in diving (decompression sickness). In all of

these mentioned areas the formation of bubbles and blisters in matrices after a sudden

pressure drop is a well-known phenomenon and part of important research work. It is

known that the phenomenon is based on the supersaturation of gas in the polymer and the

accompanied thermodynamical imbalance, hence an increase in free energy in the system

at rapid decompression. The expansion of gas leads then to the mostly unwanted blistering

effect.

Only the exact mechanisms of the formation of bubbles are still topic of current research.

Especially at HPP of food, where pressures of up to 6,000 bars are reached at moderate

temperatures only very few data is available. Considering relevant literature it could be

shown that especially the solubility but also diffusion and permeability coefficient of

gases in polymers under high pressures are the most important factors influencing the

formation of bubbles. Therefore the first part of the review is dealing with the solution

and transport properties of gases in polymers under pressure. In a second part it is

concentrated on the theory of bubble nucleation and the influence of other parameters like

material properties, gas nature and process conditions (e.g. temperature, decompression

rate etc.) on bubble density and/or bubble growth. Scientific gaps, especially concerning

very high pressure conditions are highlighted.


22

The theory of decompression failure in polymers during the

high-pressure processing of food

JULIA STERR1, BENEDIKT STEFAN FLECKENSTEIN1, HORST-CHRISTIAN LANGOWSKI1,2

1Chair for Food Packaging Technology, Technische Universität München, Center of Life and Food Sciences Weihenstephan, Freising-Weihenstephan, Germany

2Fraunhofer-Institute for Process Engineering and Packaging, Freising, Germany

In Food Engineering Reviews. Published online: 13. October 2017. DOI:

10.1007/s12393-017-9171-9

ABSTRACT

The occurrence of blistering and the formation of bubbles in matrices after a sudden

pressure drop is a well-known phenomenon in many fields, including in the petroleum

industry (“explosive decompression failure”), in diving (decompression sickness), in the

infrastructure of hydrogen fuel cells, in the foaming of polymers, and in the high pressure

processing of food (HPP). This usually undesirable effect is caused by the increased

absorption of gas in the polymer under high pressure conditions and the subsequent

supersaturation and increase in free energy on rapid pressure release. The exact

mechanisms of the resulting expansion of gas, and hence the formation of bubbles, are

not fully understood. Regarding the high pressure processing of food where pressures of

up to 6,000 bars are reached at moderate temperatures, little information is available about

the key factors involved in decompression failure. This review summarises results and

findings from relevant research areas to understand polymer decompression failure. The

first part of this review describes the transport properties of gases in polymers under high

pressure (sorption and desorption, diffusion coefficient and permeability coefficient). The

second part focuses on damage mechanisms and discusses parameters such as material

properties, the nature of the gas and process conditions (e.g. temperature, decompression

rate). Knowledge gaps and proposed research are highlighted.


23

Content Publication I

2 The theory of decompression failure in polymers during the high-pressure processing

of food ............................................................................................................................. 21

2.1 Abbreviations ................................................................................................... 25

2.2 Introduction ...................................................................................................... 26

2.3 Solution, diffusion and permeation of gases in polymers under pressure ....... 27

2.3.1 The solubility of gases in polymers under pressure ................................. 27

2.3.1.1 Influence of pressure on solubility .................................................... 30

2.3.1.2 Influence of temperature on solubility .............................................. 31

2.3.1.3 Influence of the nature of the gas on solubility ................................. 32

2.3.1.4 Influence of polymer structure, filler particles, plasticizers and

composites on solubility ...................................................................................... 33

2.3.2 Swelling and volume changes of polymers due to gas sorption ............... 35

2.3.2.1 Influence of temperature on the swelling behaviour of polymers ..... 36

2.3.2.2 Influence of pressure on the swelling behaviour of polymers .......... 37

2.3.2.3 Correction of solubility calculation due to swelling behaviour ........ 37

2.3.3 Plasticisation and changes in the glass transition temperature ................. 38

2.3.4 The diffusion coefficients of gases in polymers under pressure .............. 39

2.3.4.1 Differentiation between sorption diffusivity (Ds) and desorption

diffusivity (Dd) .................................................................................................... 39

2.3.4.2 Influence of pressure and temperature on the diffusion coefficient .. 40

2.3.5 Permeation of gases through polymers at high pressures ......................... 43

2.4 Theory of the formation and nucleation of bubbles ......................................... 47

2.4.1 Definition and formation of bubbles ........................................................ 47

2.4.2 Factors affecting the formation of bubbles in polymers ........................... 49


24

2.4.2.1 Influence of the nature of the gas and the transport properties on bubble

formation 49

2.4.2.2 Influence of the material properties on bubble formation ................. 50

2.4.2.3 Influence of pressurising conditions ................................................. 52

2.4.3 Bubble formation in packaging used for HPP .......................................... 57

2.5 Conclusion ....................................................................................................... 58

2.6 Publication bibliography .................................................................................. 61


25

2.1 Abbreviations

APET Amorphous polyethylene terephthalate BD Bubble density = bubbles/blisters per volume of the matrix BS Bubble size, volume or diameter DCS Decompression sickness EPDM Ethylene propylene diene, M-class GR Bubble growth rate HPP High pressure processing MAP Modified atmosphere packaging NBR Acrylonitrile butadiene rubber PA11 Polyamide 11 PC Polycarbonate PEEK Poly(ether-ether-ketone) PE-HD Polyethylene high density PE-LD Polyethylene low density PET Polyethylene terephthalate PGA Poly(glycolic acid) PLLA Poly L-lactic acide PLG Copolymer of D,L-lactide and glycolide (varying content) Pmax Maximum applied pressure in the system PMMA Poly(methyl methacrylate) PP Polypropylene PS Polystyrene PSS Supersaturation pressure PSU Polysulfone PTFE Polytetrafluorethylene (Teflon) PVAc or PVA Poly(vinyl acetate) PVC Poly(vinyl chloride) PVDF Poly(vinylidene fluoride) scrCO2 Supercritical carbon dioxide SRR Supersaturation ratio STP Standard temperature (273 K) and pressure (1.013 bar) TiO2 Titanium dioxide Tg Glass transition temperature Tm Melting temperature VMQ Vinyl methyl polysiloxane XDF Explosive decompression failure


26

2.2 Introduction

High pressure processing (HPP) is an innovative and increasingly used technology for

prolonging the shelf life of temperature-sensitive food. In industry pressures up to 6,000

bars are reached and temperatures between 5 and 40°C are common. Under these

conditions most bacteria and fungi are inactivated. To avoid recontamination via cost and

time consuming aseptic filling after HPP, the food is packed and sealed before treatment.

In many cases vacuum packaging is used. Increasingly, however, modified atmosphere

packaging (MAP) with gases such as nitrogen (N2), carbon dioxide (CO2) and oxygen

(O2) is being utilized in order to use the synergistic effect of high pressure and CO2 against

certain enzymes and microorganisms (Al-Nehlawi et al. 2014, Corwin and Shellhammer

2002).

The negative side effect of using gases in the headspace of packaging is the expansion of

gases during (fast) pressure release and the accompanying formation of bubbles, blisters

or foam like structures in the polymer and food (Bull et al. 2010, Fairclough and Conti

2009, Fleckenstein et al. 2014, Götz and Weisser 2002, Koutchma et al. 2009, Masuda et

al. 1992, Richter et al. 2010, Richter 2011, Sterr et al. 2015a). This phenomenon of

decompression failure is also well known in other fields such as decompression sickness

(DCS) for divers at pressures up to 10 bars or as “explosive decompression failure” (XDF)

in the petroleum industry where polymers are used as seal barriers in flexible pipes for

petroleum transportation processes with CO2 (P< 1,000 bars and 70 <T <130°C) (Baudet

et al. 2009, Boyer et al. 2007, Grolier and Boyer 2007). A similar effect also arises in the

area of hydrogen fuel cells working with pressures up to 1,000 bars (Barth et al. 2013,

Koga et al. 2013). In the food packaging industry the formation of bubbles and cracks is

an undesirable effect as it negatively affects the overall packaging integrity, barrier

properties or leads to film delamination (Fleckenstein et al. 2014, Götz and Weisser 2002,

Sterr et al. 2015a). The exact mechanism of the decompression failure is still not fully

understood and is the subject of ongoing research. This is particularly so for food

packaging applications. This review therefore collates data and information from other

research areas and considers their relevance to HPP. The first part of this review discusses

the dissolution, diffusion and permeation of gases in and through polymers under high

pressure conditions. This review focuses primarily on gases appearing in MAP including

O2, N2 and CO2 and on common polymers used in the food industry such as polyethylene


27

(PE), polyethylene terephthalate (PET), polypropylene (PP), polyamide (PA) and

polystyrene (PS). Other gases and polymers are considered only if important for further

understanding of decompression failure induced by high pressure. Differentiation is made

between results derived from experiments working with high gas pressures (i.e. two-

phase systems) and experiments working with high isostatic pressures, where the pressure

is transmitted by an incompressible fluid (i.e. three-phase systems). The second part of

the review discusses the theory of bubble formation and options for reducing

decompression failure. The transport properties of gases in polymers, the mechanical

properties of polymers and the process conditions such as the temperature and the

pressure profile are taken into account.

2.3 Solution, diffusion and permeation of gases in polymers under

pressure

The formation of bubbles and the amount and growth rate of bubbles are strongly

dependent on the amount of gas that is able to dissolve in the polymer (solubility or

sorption coefficient S) and the speed at which the gas diffuses through the polymer film

(diffusion coefficient D) (Flook 2000). Different methods have been used for measuring

the solubility, diffusion and permeability coefficient of gases through polymers under

pressure (Flaconnèche et al. 2001b). These differing measurement methods can result in

different outcomes. Further discrepancies arise in the application of “open” and “closed”

pressure cells, with a dynamic gas flow with unlimited access of the polymer to gas in the

former and a limited access to gas in the latter. The applied pressure in open systems gives

a constant high partial pressure difference on the upper and bottom side of the polymer

(flow-method), whereas the applied pressure in closed systems is uniform over the whole

sample (hydrostatic) which means that only the volume of the sample changed during

pressurization but not its shape. This hydrostatic system is also applied during the high

pressure processing of food and is therefore of special interest.

2.3.1 The solubility of gases in polymers under pressure

In general the solubility S, also called sorption coefficient, describes the amount (e.g.

concentration c) of a substance (gas) in a matrix (polymer) under equilibrium conditions

at an applied partial pressure p (Barrer 1951). An explanation for variable results in the


28

literature may be the different methods of determining and expressing solubility (Battino

and Clever 1966). In this review the solubility is described as the weight of gas relative

to the weight of the polymer � ( ��) (��, as is usual in the literature. Partial pressures

above the solvent differing from the standard pressure of 1.013 bar may be corrected by

Henry’s law (Equation 2-1), a simple relationship that applies when there is

proportionality between the molar fraction xg [mol/mol] of gas diluted in the liquid

(described by the Henry volatility constant K [bar]) and its partial pressure pg in gas phase

(Prausnitz et al. 1999, Sander 2015). There are also different ways of defining Henry’s

constant, as described in the work of Sander (2015).

Equation 2-1

�� = � ∗ ��

Under low pressure conditions and at low gas concentrations Henry’s law is independent

of the pressure and concentration. However, there are some restrictions to the application

of Henry’s law at high concentrations of the solutions (the concentration of the substance

or gas should not exceed a molar fraction of 3%) and at high pressures (the partial pressure

should not exceed 5 to 10 bars), and when there are reactions of the solvent with the gas

(Prausnitz et al. 1999). For real gases at high pressure, the partial pressure in Equation

2-1 must be replaced by the partial molal fugacity fi of the dissolved gas at the solvent

total pressure (Wedler 1997). The fugacity � describes the deviant behaviour of a real

gas to the ideal gas as a function of temperature and pressure and is defined by the

dimensionless fugacity coefficient ϕ and the partial pressure pi (of an ideal gas) as shown

in Equation 2-2.

Equation 2-2

! = " ∗ �!

At high pressure the fugacity coefficient differs from one. Table 2-1 shows values of the

fugacity coefficient of N2, O2 and CO2 in water (Atkins and Paula 2006, Wedler 1997,

Enns et al. 1965, Ludwig and Macdonald 2005, Spycher et al. 2003) and of CO2 and CH4

in PE (Sarrasin et al. 2015) at pressures up to 5,000 bars. The low fugacity values of CO2

in PE were explained by the polarity of the molecules. This affects the attractive

interactions, due to the fact that CO2 is considered as a quadrupole. The CH4 fugacity in

PE decreases only slightly up to 300 bars and increases to ~2.9 at 2,000 bars. The


29

coefficients of CH4 and CO2 in PE and of CO2 in water presented in Table 2-1 were

estimated based on a graph displayed in the work of Sarrasin et al. (2015) and Spycher et

al. (2003).

Table 2-1 Fugacity coefficients of N2, O2 and CO2 in water and of CO2 and CH4 in PE

P

[bar]

φ [-] N2 in H2O

at 273 K

φ [-] O2 in H2O

at 277 K

φ [-] CO2 in H2O

at 300 K

φ [-] CO2 in PE

at 333 K

φ [-] CH4 in PE

at 333 K

Atkins and Paula 2006; Sarrasin et al. 2015; Wedler 1997

Enns et al. 1965; Ludwig and Macdonald 2005

Spycher et al. 2003

Sarrasin et al. 2015;

Sarrasin et al. 2015;

1 0.99955 ~1 ~1 10 0.9956 ~0.97 100 0.9703 ~0.49 ~0.7 300 1.52 ~0.24 ~0.79 600 ~0.21 880 ~1 1,000 1.839 4.0 ~0.4 ~1.14 1,330 6.34 2,000 16.0 ~2.9 5,000 1,041

Different models exist for explaining diffusion and sorption processes of gases in

polymers. Two models will be discussed here: The free volume theory and the dual

sorption model. The free volume model is based on the assumption that the diffusing

molecules are located in micro voids (free volumes) of the polymer matrix and movement

of the molecules only takes place via statistical fluctuations of these voids by chain

motion. For semi-crystalline polymers the theory predicts that it is largely the amorphous

phase of the polymer or voids that are penetrable by the gas (Baudet et al. 2009, Michaels

and Bixler 1961, Scheichl et al. 2005). Boyer et al. (2006) assumed that at higher

pressures sorption may also take place in interstitial sites deeper in the polymer, such as

in the crystal phase. The free volume model has been extensively described and reviewed

elsewhere (Budd et al. 2005, Duncan et al. 2005, Fujita 1961, Mercea 2008, Vrentas et

al. 1993, Vrentas and Duda 1977). Grolier and Boyer (2007) split the pressure dependence

of CO2 sorption into three parts. At low pressures, the sorption of gas on the surface and


30

in the amorphous phase is an exothermic process. Secondly, at higher pressures, CO2 also

penetrates voids and interstitial regions in an endothermic process. At pressures above

300 bars the polymer acts as a pseudo-homogeneous phase saturated with supercritical

CO2. In scientific work concerning the influence of high hydrostatic pressure on the

diffusion process it has become clear that the free volume is compressed by the pressure

and therefore gas transport is reduced during this process (Fleckenstein et al. 2014,

Richter et al. 2010, Richter 2011, Sterr et al. 2015b). However, plasticization effects may

have an opposite effect on the transport properties of gases.

The dual sorption model assumes two types of sorption and solving molecules (Hilic et

al. 2001, Muth et al. 2001, Vieth et al. 1976, Vieth 1991). Firstly, at the Langmuir type

absorption sites in micro voids, molecules are completely immobilized, and secondly, in

accordance with Henry’s law, small molecules with weak interactions are dissolved in

the polymer matrix. Diffusion only occurs for the molecules dissolved in Henry’s mode

(Tsujita 2003). The experiments of Hilic et al. (2001) show good agreement of the

solubility of N2 in PS up to 600 bars with the dual sorption model, whereas Chang et al.

(1998) found the model to be inadequate for pressures exceeding 100 bars for CO2

sorption in glassy polymers such as PC (35 to 50°C, 300 bars). Both were working with

static gas pressures.

Several parameters that influence the sorption of gases in polymers have been identified

including the temperature, pressure filler particles in the polymer, stiffness and the

crosslinking density of the polymer chains (see sections below). Bonavoglia et al. (2006)

also included the glass transition temperature and degree of crystallinity as parameters.

Illustrative values for gas solubility in polymers at high pressure are listed in Table 2-2.

More data on the solubility of gases in polymers under high static and dynamic pressures

can be found in the work of Boyer et al. (2007), Briscoe and Mahgerefteh (1984),

Lundberg et al. (1969) and Naito et al. (1991). For data on the transport properties at

atmospheric pressure the reader is referred to the permeation and diffusion data of other

authors (including Brandrup et al. 1999, Lewis et al. 2003).

2.3.1.1 Influence of pressure on solubility

It is generally observed that the solubility of gases in polymers increases with pressure

(Areerat et al. 2002, Fairclough and Conti 2009, Kulkarni and Stern 1983, Lei et al. 2007,

Ru-Ting and Xing-Yuan 2015, Sato et al. 1999, Tang et al. 2004b, Tang et al. 2007).


31

However, a few authors have observed a saturation behaviour of gases in polymers at

elevated gas pressures (Boyer et al. 2011). For example, Muth et al. (2001) observed a

flattening behaviour for CO2 sorption in poly(vinyl chloride) (PVC) at 400 bars, but did

not consider it as saturation behaviour. Briscoe and Zakaria (1991) found that the curve

of mass uptake of CO2 in a silicone elastomer at 42°C flattened from 110 bars up to 300

bars. Similar behaviour was observed by Spyriouni et al. (2009) for CO2-PS systems (35

<T <132°C; 1< P< 300 bars). The levelling-off seemed to be more pronounced at the

lower temperatures. Reasons for a solubility limit might reflect a decrease of the free

volume, in addition to decreased chain motion. Sarrasin et al. (2015) observed an

increasing but limited uptake of CO2 in the amorphous PE phase (T = 60°C; 200 < Pgas

< 1,000 bars). On the other hand, a decreasing solubility coefficient was observed for the

same system under the same conditions when high hydrostatic pressures were applied

(i.e. constant gas concentration).

Fairclough and Conti (2009) calculated the relative increase in solubility of N2 in PP for

gas pressures up to 7,000 bars at different temperatures on the basis of the ratio of the

solubility at a given pressure to the solubility of the gas at liquid densities. They found

that the solubility increases 50 fold at 20°C and 7,000 bars. However, no comparison with

empirically collected data was undertaken. No other experimental data on the solubility

of gases in polymers under isostatic pressures have been reported. A thermodynamic

calculation developed by Klotz (1963) predicts that the solubility of O2 in sea water is

almost independent of the depth up to 1,000 m below the surface (~100 bars), whereas

the N2 solubility decreases 5 to 6% at 1,000 m depth. Wiebe and Gaddy found that the

solubility of CO2 in water increases by a factor of 4 (at 50°C) to 7 (at 100°C) when

increasing the pressure from 25 to 700 bars (Wiebe and Gaddy 1939, 1940). No other

investigations of the influence of isostatic pressures up to 6,000 bars or higher on the

solubility of gases in polymers could be found in the literature.

2.3.1.2 Influence of temperature on solubility

At low pressures and temperatures up to 100°C the solubility of gases in common solvents

decreases with increasing temperature. At higher temperatures, the solubility can increase

with temperature (Prausnitz et al. 1999). At high pressures, two different effects of

temperature on the solubility of gases in polymers are described in the literature. At

constant pressure, increasing temperature decreases the solubility of gases in polymers


32

(Battino and Clever 1966, Boyer et al. 2011, Hilic et al. 2001, Sato et al. 1999, Sato et al.

2000, Tromans 1998). Areerat et al. (2002) measured a negative slope for CO2 solubility

in PP and PE-LD/TiO2 systems with increasing temperature from 150 to 200°C and at

constant pressures up to170 bars. Similar results were obtained by Lei et al. in a CO2–PP

system with increasing temperature (40< T< 120°C and 160 < T< 210°C) and at constant

gas pressures up to 250 bars. When compared to the rubbery state, these authors observed

a higher gas solubility in polymers in the molten state (>160°C). A larger fraction of

available amorphous regions was assumed to explain this observation. Thus, gas

solubility increases in the 120 to 160°C transition region (Lei et al. 2007). Hilic et al.

(2001) presumed that a negative slope of the temperature dependence of the gas solubility

should be commonly observed in binary systems comprising an amorphous polymer and

light gases such as N2, O2 and Ar at temperatures below the glass transition temperature

Tg. “Reverse solubility”, i.e., when the gas solubility increases with temperature, is

observed for gases with low critical temperatures (e.g. N2 and H2; Sato et al. 1999) but

only at high temperature. However, this effect has been found for N2 and He in

polytetrafluorethylene (PTFE) even at low temperatures (T< 80°C) and 380 bars (Briscoe

and Mahgerefteh 1984). Flaconnèche et al. (2001a) measured a divergent solubility

dependence on temperature. In PE they did not find a measurable influence of temperature

on the solubility of CO2, N2 or He. CH4 solubility in PE showed reverse solubility in the

80°C to 40°C range, whereas CO2 solubility in polyamide (PA11) and poly(vinylidene

fluoride) (PVDF) increased with decreasing temperature while it decreased for N2 and

He. The solubility of CH4 in PA11 and PVDF seemed to be unaffected by temperature.

A possible explanation for some of these conflicting results could be the different

measurement methods used at high pressures and the inclusion of swelling effects (Lei et

al. 2007). Additionally, adiabatic heating was not taken into account in any of the relevant

publications. In the case of HPP applications, heat transfer to the pressurizing medium

should also be considered.

2.3.1.3 Influence of the nature of the gas on solubility

Gas solubility is strongly dependent on the gas-polymer system properties. Experiments

with silicone elastomers showed that the mass uptake of N2 into the material is 6 to 7

times less than the CO2 uptake observed at 42°C and pressures up to 220 bars (Briscoe

and Zakaria 1990b). Flaconnèche et al. (2001a) measured the solubility of CO2, CH4, Ar


33

and N2 in PE, PA11 and PVDF (40< T< 80°C and P< 120 bars). They observed the highest

solubility values for CO2, followed by CH4, Ar and N2. In other studies (40 <T < 160°C

and 100 < P < 220 bars) it was confirmed that CO2 has 4 to 16 times higher solubility

than N2 in PE-HD, PP, PS and also a silicone elastomer (Briscoe and Zakaria 1990b, Sato

et al. 1999, Sato et al. 1996). These results at high gas pressures are similar to the

solubility behaviour of gases under atmospheric conditions (Michaels and Bixler 1961).

The presence of additional gases may alter the transport behaviour of gas. This might be

of importance for the high pressure processing of food, if modified atmosphere packaging

is used. This phenomenon is also observed at atmospheric pressure and it has been shown

that the solubility of N2 and O2 in PE is affected by the presence of CO2, whereas the

mechanisms of solubility and diffusion of CO2 seemed to be unaffected by other gases

(Pino et al. 2005). Comparable results were found by Lewis et al. (2003) concerning the

modified transport properties of N2 in PET in the presence of CO2 or O2. A decreasing

solubility of CO2 in a silicone rubber (poly(dimethyl) siloxane) at pressures up to 60 bars

due to the presence of N2 was measured by Jordan and Koros (1990). Sarrasin et al. (2015)

found that the solubility coefficients of H2S and CH4 in PE were not altered by the

presence of the other gas for total gas pressures of 2,000 bars.

2.3.1.4 Influence of polymer structure, filler particles, plasticizers and composites on

solubility

The maximal amount of gas taken up by a polymer under elevated gas pressures is also

influenced by the structural properties of the polymeric material such as the degree of

crystallinity, density and molecular weight (Sato et al. 1999). Other studies have shown

that higher stiffness and an accompanying lower amorphous content reduces the amount

of dissolved gas (CO2, N2, Ar, etc.) in polymers such as PE and elastomers at high static

gas pressures (Briscoe and Kelly 1996, Briscoe and Zakaria 1990b, Flaconnèche et al.

2001a). Under atmospheric conditions, a linear correlation between the solubility of

thirteen different gases in PE and the amorphous content was found by Michaels and

Bixler (1961).

In the studies of Briscoe and Zakaria (1992) it was shown that elastomers with filler

particles could take up more gas at high pressures than the pure polymer, due to the weak

internal matrix-filler bonds, except when a gas such as CO2 is able to rupture the internal

interfaces. In general, the affinity of the diffusing molecules to the filler particles and


34

hence the solubility of the gas in the filler particles is of importance for reasoning this

behaviour (Duncan et al. 2005). Briscoe et al. (1994) presents a good overview of the

effect of filler particles on gas solubility. Areerat et al. (2002) pointed out the different

ways for examining gas solubility in polymers with filler particles. The “apparent”

solubility, which is defined as the weight of dissolved gas per unit weight of the

composite, was compared to the “true” solubility, which is defined by the weight of

dissolved gas per unit weight of polymer. Studies on the solubility of CO2 in a blend of

PE and titanium dioxide (TiO2) at high pressures (150 < T< 200°C and P= 150 bars)

showed that the apparent solubility decreases with increasing TiO2 content, whereas the

true solubility stays constant, even when the TiO2 content changes. No information was

given about the solubility of CO2 in TiO2. Yamabe and Nishimura (2010) tested the

influence of carbon black and silica filler particles in rubbery polymers on decompression

failure (T = 30°C, P = 100 bars) and observed that composites without particles and with

silicon particles take up less H2 than with carbon black particles. This is explained by the

ability of carbon black to absorb H2, whereas silica does not absorb H2. Flaconnèche et

al. (2001a) observed that the incorporation of different weight fractions of plasticizer (n-

butyl-benzene-sulphonamide) had only a minor influence on the solubility of CH4 in

PA11 (T = 120°C and P = 40 bars). The solubility of CO2 in PA11 under the same

conditions was slightly increased by incorporation of 29.5% plasticiser.

Table 2-2 Effect of temperature (T) and pressure (P) on the solubility of gases in various polymers

Gas Polymer T [°C] P [bar] Solubility / Sorption [g(gas)/ g(polymer)]

Reference

Experimental temperatures below the glass transition temperature Tg of the polymers N2 PS 40

60 80

161 (1) 165 (1) 175 (1)

0.008 0.006 0.006

(Sato et al. 1999)

N2 PS 40 60 80

700 (1) 500 (1) 620 (1)

0.026 0.021 0.018

(Hilic et al. 2001)

CO2 PVC 50 400 (1) 0.12 (Muth et al. 2001) CO2 PVC

PC PMMA

25 62 (1) ~0.08 ~0.12 ~0.25

(Berens et al. 1992)

CO2 PSU PC

40 400 (1) 0.12 ~0.14

(Tang et al. 2004a, Tang et al. 2004b)

Process temperatures ranging between Tg and Tm of the polymer


35

O2 PE-LD PP

25 50 (2) 0.003 0.003

(Naito et al. 1991)

CO2 PE-LD PP

25 50 (2) 0.023 0.029

(Naito et al. 1991)

CH4 PE 60 2,000 – 5,500 (3)

~0.03 – 0.04 (Sarrasin et al. 2015)

CO2 PE 60 500 (3) ~0.05 (Sarrasin et al. 2015) CO2 PE 20 23(1) 0.012

(Kulkarni and Stern 1983)

CO2 PE-MD PE-LD

40 40(1) 0.047 0.07

(Flaconnèche et al. 2001a)

N2 PE-LD PA11

70 100(1) 0.004 0.003


CO2 PE-MD 60 350(1) ~0.08 (Boyer et al. 2007) CO2 PP 40 100(1) 0.0201 – 0.0542 (Lei et al. 2007) CO2 PP 100 120(1) ~0.13 (Ru-Ting and Xing-

Yuan 2015) CO2 PET 80

120 300(1) ~0.030

~0.033 (Schnitzler and Eggers 1999)

(1) Gas pressure (2) Dynamic pressure. Pressure difference between upstream and downstream gas flow ΔP(dyn) (3) Theoretical approach, computational simulations and/or calculations

2.3.2 Swelling and volume changes of polymers due to gas sorption

The determination of the solubility of gases in polymers is often measured by the relative

volume change or the swelling of the polymer caused by the uptake of gases (Boyer and

Grolier 2005b, Kazarian 2000). Concerning the measurement of solubility via the volume

change of a polymer, the opposite effect of hydrostatic compression has to be taken into

account under high pressure conditions. Ultimately the measured volume is a result of

several factors such as the solubility of the gas (increasing solubility and amount of

dissolved gas leads to increasing volume), the maximum applied pressure, the

temperature and/or thermal expansion and whether the polymer has a rubbery, glassy or

semi-crystalline structure (Bonavoglia et al. 2006, Briscoe and Zakaria 1991, Hilic et al.

2001, Jordan and Koros 1990). Increases in the polymer volume due to swelling range

between 1% for N2 in PS to 130% for CO2 in a rubbery polymer (Table 2-3). The negative

(relative) volume change of the PE amorphous phase is about 6% in the absence of gas,

namely neglecting swelling, for pressures up to 2,000 bars (Sarrasin et al. 2015). The


36

methods of measuring the gas solubility and the associated swelling have been reviewed

by Grolier and Boyer (2007).

Table 2-3 Effect of temperature (T) and pressure (P) on the relative volume change (RVC) of gases in various polymers

Polymer Gas T [°C] Gas P [bar] RVC [%] Reference PET CO2 60 300 +2.4 (Schnitzler and

Eggers 1999) PET CO2 40 300 +2.5 PC CO2 40 300 +9.5 PS CO2 35 100 +11 PMMA CO2 35 100 +17 Rubbery polymer

CO2 35 300 +130 (Chang et al. 1998)

PE CO2 80 1,000 +12 (Sarrasin et al. 2015)

PS N2 40 40 80

266 695 625

+1.1 +2.2 +0.7

(Hilic et al. 2001)

PS CO2 45 90

102 246

+9.4 +13.3

2.3.2.1 Influence of temperature on the swelling behaviour of polymers

Schnitzler and Eggers (1999) found that the swelling of PET decreased with increasing

temperature under low temperature conditions (40< T< 60°C), namely at temperatures

less than Tg, due to the lower CO2 density which resulted in a decreased amount of

dissolved CO2. On the other hand, the swelling increased with increasing temperature at

higher temperatures (70 < T< 120°C), namely temperatures below melting temperature

Tm, due to the higher chain mobility. These results were confirmed by Lei et al. (2007)

for a CO2-PP system up to 100 bars and by Spyriouni et al. (2009) for a CO2-PS system,

in particular for pressures below 100 bars. Also, Hilic et al. (2001) reported that the

relative volume change in PS due to dissolution of N2 and CO2 decreased with

temperature at temperatures below Tg. It was also presumed that the relative volume

change is more important below Tg, because of the plasticisation effect of the solute, and

thus solubility and volume changes are less pronounced above Tg. Schnitzler and Eggers

(1999) assumed that the amount of swelling increases significantly on exceeding the glass

transition temperature of the polymer. When the temperature approaches the melting

temperature Tm, the amount of swelling increases with temperature. The authors assumed

that the compression of the polymer by pressure is negligible. No published data were


37

found concerning the temperature dependent swelling behaviour under high hydrostatic

pressure with consideration of compression effects.

2.3.2.2 Influence of pressure on the swelling behaviour of polymers

Sarrasin et al. (2015) found that the swelling of the amorphous phase in PE by CO2

reaches a maximum of about 8% (relative volume change) at 60°C and about 12% at 80°C

for pressures up to 1,000 bars. The limited swelling ability was explained by the

antagonistic effect of gas uptake and a compression effect that reduces the free volume

and hence the sorption process. Likewise in the studies of Hilic et al. (2001) (N2-PS

system at 40< T< 80°C, P = 700 bars and CO2-PS system at 65< T< 130°C, P = 450 bars),

in the work of Spyriouni et al. (2009) (CO2-PS system at 35< T< 130°C, P = 300 bars)

and in the experiments of Schrittesser et al. (2016) (CO2-elastomer system up to 150 bars),

the relative volume changes seemed to flatten with increasing gas pressures. However, no

maximum was detected. The levelling off appears to be more pronounced at lower

temperatures (32 and 50°C). On the contrary, Lei et al. (2007) noted that swelling of PP

due to CO2 sorption (40< T< 210°C) increases with pressure at higher pressures up to 250

bars.

2.3.2.3 Correction of solubility calculation due to swelling behaviour

Lei et al. (2007) and Sato et al. (2001) included the polymer swelling in their gas solubility

estimations. The corrected solubility values are higher by a factor of 7 to 20% compared

to the values obtained when neglecting swelling (e.g. using gravimetric measurements)

and the factor increases with temperature (Sato et al. 2001). In contrast, the swelling-

corrected values of Lei et al. (2007) were up to 160% higher and increased with pressure.

Bonavoglia et al. (2006) corrected the values of swelling as a function of dissolved CO2

concentration and noted that the curve in general can be divided into three regions. In the

first region, gas sorption occurs without swelling of the polymer. In the second region

swelling and gas sorption go hand in hand and in the third region swelling can be

neglected again. Nilsson et al. (2013) noted that the calculation of solubility is

independent of the swelling behaviour, because the solubility is linearly related to the

density.


38

2.3.3 Plasticisation and changes in the glass transition temperature

Swelling does not merely change the volume of a polymer, it also increases the free

volume and chain mobility (Chiou et al. 1985b, Kazarian 2000, Tang et al. 2004b). This

phenomenon, called plasticisation, has an influence on the solubility and the rate of

diffusion (See section Diffusion), which may then become a function of concentration and

time and show non-Fickian behaviour (Klopffer and Flaconnèche 2001). The antagonistic

effects of compression and swelling (or plasticisation) on the molecular chain motion also

have to be considered here (Briscoe and Zakaria 1991). The plasticisation is linked to

changes in the glass transition temperature Tg. Gases with higher critical temperatures

such as CO2 and CH4 are more condensable and thus have a strong plasticising effect with

an accompany effect on the Tg of the polymer (Bonavoglia et al. 2006, Berens et al. 1992,

Chiou et al. 1985a, Jordan and Koros 1990). Plasticisation by these gases may dominate

the compression effect, whereas for gases with low sorption behaviour such as N2 and He

the compressing effect may dominate. Another explanation was proposed by Bos et al.

(1999) who predicted that polarisable CO2 may interact with polar groups of the polymer

and therefore show higher plasticisation tendency.

The influence of the applied pressure and the additional sorption of gas on Tg have been

analysed by several authors (Berens et al. 1992, Bonavoglia et al. 2006, Boyer and Grolier

2005a, Grolier and Boyer 2007, Schnitzler and Eggers 1999, Tang et al. 2007, Wang et

al. 1982, Wessling et al. 1994, Zhang and Handa 1998). All concurred that increasing gas

pressure reduces Tg via the increased gas sorption and the resulting plasticising effects.

Examples of values calculated using a model developed by Chow (1980) for predicting

the pressure induced changes in Tg show that N2 is able to reduce the Tg of PS by 47°C

(at 40°C and pressures up to 700 bars), whereas CO2 can depress the Tg of PS by 80 to

100°C at 250 bars and a temperature of 65°C (Boyer and Grolier 2005a). Similar results

were obtained by Bonavoglia et al. (2006), Berens et al. (1992) and Chiou et al. (1985a)

for common glassy polymers such as PMMA, PC, PVA, PVDF, PET and PVC.

Only a few reports support the idea that the Tg may also increase with pressure, in

situations when the polymer matrix compression effect due to pressure dominates the

plasticisation effect of the gas (Briscoe et al. 1994, Campion and Morgan 1992, Wang et

al. 1982). Additionally, it is assumed that this is more common in elastomers than in

glassy polymers. In experiments working at hydrostatic pressures up to 10,000 bars (water


39

or petroleum ether as pressure transmitting fluid), Paterson (1964) found that the Tg of

several rubbers increased by ~16°C per 1,000 bars, but only at pressures above 1,000 to

5,000 bars and when the Young’s Modulus increased 1,000 fold. At lower pressures, the

Young’s Modulus only increased by a factor of 2-3. A good but older overview of

pressure induced changes on the mechanical properties and of the pressure-induced

increase in Tg of polymers can be found in the report by Jones Parry and Tabor (1973).

Readers can find additional information on sorption-induced plasticisation and

crystallisation in the work of Fleckenstein et al. (2014) and Kazarian (2000). For detailed

data on the glass transition temperatures of polymers, readers are referred to the

publication of Brandrup et al. (1999).

2.3.4 The diffusion coefficients of gases in polymers under pressure

The diffusion coefficient D describes the mobility of molecules passing through a matrix

by random molecular motion and is defined by the root mean square displacement of the

molecules per time [m²/s] (Crank 1975). The diffusion coefficient is highly dependent on

the gas concentration and at high pressure, D increases with gas concentration (Briscoe

et al. 1994, Flaconnèche et al. 2001a, Kulkarni and Stern 1983, Naito et al. 1991). This

correlation is enhanced by the swelling of the polymer and increased chain motion due to

the plasticisation effect at high gas concentrations and high pressures. It is reduced by

compression of the matrix when applying very high pressures. Table 2-4 shows values

for the diffusion coefficient, sorption diffusivity and desorption diffusivity of gases in

polymers at high pressures.

2.3.4.1 Differentiation between sorption diffusivity (Ds) and desorption diffusivity (Dd)

Studies on the mechanisms of bubble formation require knowledge of how fast the gas

enters the polymer during pressure build up and the pressure holding time and also how

fast the gas can escape from the polymer during pressure release. In ideal systems, when

D is constant, the sorption and desorption behaviour is symmetric. However, Duncan et

al. (2005) postulated substantial asymmetry between gas uptake and release after pressure

treatment and so additional differentiation between the sorption diffusivity Ds and

desorption diffusivity Dd in polymers is necessary. In their work, they postulated a marked

hysteresis if the “absorbed molecules are strongly bound in the polymer”. Confirming

results were reported by Lorge et al. (1999) for a CO2-PVDF system at 80°C and gas

pressures up to 300 bars. They reported that gas desorption was only detectable at pressure


40

below 150 bars. Other studies have shown the hysteresis effect more often for polymers

in the glassy compared to the rubbery state (Fleming and Koros 1986). The hysteresis

effect may also reflect the swelling of the polymers which occurs during plasticisation,

or even crystallization effects and possible damage such as XDF during sudden pressure

release after high pressure processing.

Only a few studies have differentiated between Ds and Dd and found that desorption after

pressure treatment is slower than the sorption diffusivity, namely Ds > Dd (Muth et al.

2001, Tang et al. 2007). Tang et al. (2004b) reported contrary results showing higher Dd

than Ds values for CO2 in PSU and PC (T = 40°C, P = 400 bars). The results were

attributed to the plasticising effect of CO2 during the desorption process and the stronger

interaction between CO2 and the carbonyl group of PC during sorption (Tang et al.

2004b). Flook (2000) noted that the “gas washout” of saturated tissue takes longer than

the gas uptake to full saturation. The author was studying the transport properties of Ar,

N2 and He in human tissue in hydrostatic pressure systems for diving. Additional

information on desorption kinetics has been reported by Grandidier et al. (2015).

Concerning the temperature dependence of Ds and Dd it was found that at a given gas

pressure P the desorption diffusivity of CO2 in PSU and PVC increases with temperature,

whereas Dd decreases (Tang et al. 2007, Muth et al. 2001). Other authors calculated the

time to release a certain amount of gas from the polymer (Nilsson et al. 2013, Grandidier

et al. 2015, Fleming and Koros 1986, Tang et al. 2004b).

2.3.4.2 Influence of pressure and temperature on the diffusion coefficient

Experiments on the diffusion coefficient of O2 in seawater (35‰ salinity) show that D

increases with temperature. Below a temperature-dependent maximum pressure, the

diffusion coefficient increases with pressure. At 20°C, D reaches a maximum at 240 to

330 bars, whereas at 0°C data extrapolation suggest a maximum at 1,200 bars. Beyond

these pressures the diffusion coefficient decreases when the pressure is increased

(Ramsing and Gundersen 2000). Similar results were reported by Nilsson et al. (2013)

using a model predicting the diffusivity of gases in polymers at gas pressures up to 700

bars. The maximum D value is centred at around 50 to 200 bars. The reason for

diffusivities increasing with pressure at low pressures is assumed to reflect an increase in

the free volume due to gas sorption and swelling. The opposite effect at higher pressures

is assumed due to the dominating polymer compression effect relative to the swelling


41

effect. This correlation was taken into account by Naito et al. (1991) when including in

Equation 2-4 a compression effect due to the hydrostatic pressure given by the negative

parameter βh and a volume increase of the material given by the positive swelling factor

α. Both parameters are assumed to be dependent on the size of the diffusing molecules.

The diffusion coefficient D at higher pressures can be calculated from the diffusion

constant D0 under atmospheric conditions and the partial pressure of the gas pg (Equation

2-3) (Naito et al. 1991). Flaconnèche et al. (2001a) could not detect any pressure or

temperature dependence on the diffusion coefficient of CH4 or CO2 in PE in the 40 to 100

bars and the 40 to 80°C range. No relevant literature data was found concerning the

diffusion coefficients of gases in polymers at isostatic pressures.

Equation 2-3

# = #$ ∗ %&�'

Equation 2-4

( = () + + ,2


42

Table 2-4 Effect of temperature (T) and pressure (P) on the gas desorption and sorption diffusion coefficient (D) in various polymers

Gas Polymer T [°C] P [bar] D [m²/s] Reference

Desorption diffusivity Dd

CO2 PVC 50 400 (1) 0.24*10-11 (Muth et al. 2001) CO2 PSU 40 400 (1) 2.37*10-11 (Tang et al. 2004b) CO2 PMMA 40 60 (1) 1.20*10-11 (Tang et al. 2007) CO2 PS >>>PVC >PMMA 25 65 (1) 10-10 to 10-11 (Berens et al. 1992)

Sorption diffusivity Ds

CO2 PVC 50 400 (1) 1.36*10-11 (Muth et al. 2001) CO2 PSU 40 400 (1) 0.54*10-11 (Tang et al. 2004b) CO2 PMMA 40 60 (1) 6.57*10-11 (Tang et al. 2007) CO2 PS 80 245 (1) 2.4*10-10 (Arora et al. 1998)

Diffusion coefficient D

CO2 PET 80 120

300 (1) 1.7*10-11 5.0*10-11

(Schnitzler and Eggers 1999)

CO2 PS 100 83 (1) 1.67*10-10 (Sato et al. 2001) CO2 PBS 120 120 (1) 1.23*10-9 (Sato et al. 2000) CO2 PE-MD 60 40 (1) 1.04*10-10 (Flaconnèche et al. 2001a) CO2 PE 20 23 (1) 5,24*10-11 (Kulkarni and Stern 1983) O2 Rubbery polymer 25 110 (2) 7.5*10-11 (Naito et al. 1996)

PA11 70 – 100

40 – 100 (1) D(He)>>>D(CO2)≈D(Ar)>D(CH4)≈D(N2) (Flaconnèche et al. 2001a)

PE 40 – 80 40 – 100 (1) D(He)>>>D(CO2)≈ D(Ar)≈D(CH4)≈D(N2) (Flaconnèche et al. 2001a) (1) Gas pressure (2) Dynamic pressure. Pressure difference between upstream and downstream gas flow ΔP(dyn)


43

2.3.5 Permeation of gases through polymers at high pressures

The permeation is a product of the solubility and diffusion coefficient and therefore

involves complex interaction between several parameters. Measurement difficulties may

explain why few authors have reported data on the permeation of gases through polymers

at pressures exceeding 100 bars (Campion and Morgan 1992). Campion and Morgan

(1992) and Flaconnèche et al. (2001b) have described units for measuring the upstream

and downstream pressure difference of the gas to determine the permeability coefficient

P of pure and mixed gases (CH4, CO2, H2S) in polymers at temperatures up to 200°C and

dynamic gas pressures up to 1,000 bars. More recently Sterr et al. (2015b) presented a

new setup for fluorescence-based measurement of O2 permeation in an aqueous

environment through polymers at isostatic pressures (T< 50°C, P< 2,000 bars) and

independently controlled O2 partial pressures of around 0.21 bar. These studies showed

that the transport mechanisms at high pressures follow Arrhenius’ behaviour. A

measurable but complex influence of crystallinity on the permeability coefficient, due to

a plasticisation effect was observed by Flaconnèche et al. (2001b). Flaconnèche et al.

2001a reported that increasing the plasticiser content in PA11 increased the CH4 and CO2

gas permeation rate (T = 120°C, Pgas = 40 bars). Similar conclusions on the influence of

crystallinity were made by Sterr et al. (2015b) when reviewing previously reported data

(Richter et al. 2010, Richter 2011).

The studies of Sterr et al. (2015b) showed a negative exponential pressure dependence of

the permeability coefficient for pressures up to 2,000 bars at 23°C and a reduction of the

coefficient by a factor of 35 to 70 in PE films. However, the permeability coefficient at

around 100 bars is only reduced by a factor of 1.5 to 2. From the results of Naito et al.

(1991) there is a similar factor for the reduced permeability coefficient of O2 in PE at 100

bars (dynamic gas pressure, two-phase system). Campion and Morgan (1992) measured

a two-fold reduction for H2 in different elastomers at gas pressures up to 400 bars (T =

50°C). For gases with a low critical temperature such as N2 and He, Jordan and Koros

(1990) observed that the permeability coefficients decreased by a factor of 1.5 to 1.6 in

silicone rubbers for gas pressures increasing up to 62 bars at 35°C, whereas the

permeation increased with increasing pressure for CO2 (1.2 fold). These authors termed

the former gases “compressors” and the latter gases “plasticisers”. These findings were

confirmed by Naito et al. (1991) who reported that highly soluble gases with a large


44

molecular diameter (e.g. CO2, CH4) show a positive dependence of the permeability

coefficients with pressure, whereas less soluble gases (e.g. He, H2, N2, O2 and Ar) present

no reduction or only a low reduction in permeation with pressure in PP and PE-LD films

at 25°C and at pressures up to 100 bars. N2 molecules have a relatively large diameter,

like the first group of gases, but only low solubility and therefore does not show similar

permeation behaviour. It was also noted that the plasticising effect is higher in PP than in

PE due to the different amounts of crystalline regions in the polymer. Other measurements

of CO2 permeation in 11 glassy polymers showed that at gas pressures of 10 to 40 bars

the permeability coefficient decreases with pressure, while it rises at pressures above 50

bars. The pressure at which the minimum in permeation occurs was called the

plasticisation pressure (Bos et al. 1999). It is questionable whether this effect is valid at

pressures up to 6,000 bars, as applied in the high pressure processing of food. Probably

the compression of the matrix at high hydrostatic pressure overcomes the plasticisation

effect. Unfortunately no relevant data on gas permeation through polymers at such high

hydrostatic pressures is available.

It is also important to examine the behaviour of gas mixtures, given that the permeation

of a particular gas may be influenced by other gases (Scheichl et al. 2005). In the case of

CO2, its permeability coefficient in silicone rubber is decreased by N2 (Jordan and Koros

1990). This may reflect the compressive nature of N2. This effect also works the other

way around, with increased permeation and an increased diffusion coefficient for N2 in

the presence of CO2. In general, the effect depends on the mixing ratio of the gases.

Additional permeability coefficient data and general observations on reported values are

summarised in Table 2-5.


45

Table 2-5 Effect of maximum pressure (Pmax) and temperature (T) on the permeability coefficients (Q, P) at high pressure of gases in various polymers

Gas Polymer T [°C] Pmax [bar]

Permeability Q � ./0/1∗2∗345�

Permeability coefficient P

�./0∗[7889/]/1∗2∗345 �

Reference

Decreasing permeability coefficient with increasing pressure

O2 PE 23 1 2,000 (3)

P 1,5*10³ - 2,1*10³ 29 to 44

(Sterr et al. 2015b)

N2

He CO2

Silicone rubber: Poly(dimethyl) siloxane

35 41 (2) P 25*10³ 36*10³ 440*10³

(Jordan and Koros 1990)

CO2 PVC 50 400 (1) P 0.44*10³ (Desorption) 2.49*10³ (Sorption)

(Muth et al. 2001)

CO2 PET 80 120

300 (1) P 1.03*10³ 3.03*10³

(Schnitzler and Eggers 1999)

CO2 PE 20 23 (1) P 11*10³ (Kulkarni and Stern 1983)

CO2 PE-LD 40 40 (2) P 19*103


N2 PE-LD 70 100 (2) P 4.4*10³ N2 PA11 70 100 (2) P 0.14*10³ He Ar N2 CH4 CO2

PE PA11 PVDF

40 – 130 40 to 120 (2)

P PE > PA11 ≈ PVDF P He ≈ CO2 >> Ar> CH4 > N2 40 – 80

40 to 100 (2) P CO2 > He > CH4 > Ar > N2

40 – 130 P He > CO2 > Ar > CH4 > N2


46

CO2 He N2

Silicone rubber: Poly(dimethyl) siloxane

35 62 (2) P CO2 >> He > N2 (Jordan and Koros 1990)

N2 H2 He

Synthetic rubber (VMQ and EPDM)

25 100 (2) Q He = H2 >> N2

(Koga et al. 2013)

No or only small reduction of the permeability coefficient with increasing pressure (sparingly soluble gases)

PP PE

25 100 (2) P He ≈ H2 >> O2 ≈ Ar >> N2 (Naito et al. 1991)

Polybutadiene 25 80 (2) P He ≈ H2 >> O2 ≈ Ar >> N2 (Naito et al. 1996)

Increasing permeability coefficient with increasing pressure (highly soluble gases)

PP PE

25 100 (2) P N2O > CO2 >> CH4 (Naito et al. 1991)

Polybutadiene 25 80 (2) P N2O > CO2 > C2H4 (Naito et al. 1996)

CO2 Silicone rubber 35 41 (2) P 440*10³ (Jordan and Koros 1990)

(1) Static gas pressure (2) Dynamic pressure. Pressure difference between upstream and downstream gas flow ΔP(dyn) (3) Hydrostatic pressure


47

2.4 Theory of the formation and nucleation of bubbles

2.4.1 Definition and formation of bubbles

Several definitions of bubbles and mechanisms for their initial formation have been

reported. Lubetkin (1994) stated that bubbles are “gas surrounded by a bilayer of

surfactant in a second gas phase, and gas surrounded entirely by a liquid with or without

surfactant”. The latter form plays a significant role in HPP. It is generally accepted that

bubbles have their origin in stable gas micronuclei or form spontaneously when the

pressure drop is high enough. In theory, nanobubbles form spontaneously after

decompression on hydrophobic surfaces from gas dissolved in the liquid (Arieli and

Marmur 2011, 2013). Gas micronuclei arise from nanobubbles and visible bubbles grow

from them. Lubetkin (1994) assumed six different sources for the formation of bubbles

including homogeneous and heterogeneous nucleation, cavitation, electrolysis, Harvey

nuclei and pre-existing, stable colloidal free bubbles. Harvey nuclei are pre-existing

nuclei with a radius greater than the critical radius of curvature (Jones et al. 1999).

Electrolysis is not relevant for bubble formation in HPP. Although homogeneous

nucleation normally occurs only in single-component systems without surfaces or

particles acting as inhomogeneities, it has been postulated that it may be relevant for

systems with supersaturation ratios (SSR) in excess of about 1,000 and thus relevant to

HPP applications (Campbell 1968, Lubetkin 1994, Wilt 1986). The dimensionless SSR of

a system is defined as the ratio of the equilibrium pressure of the gas to the pressure in

the system (Wilt 1986). According to Wilt (1986), a supersaturation ratio of more than

20 (SRR> 20) is required for heterogeneous nucleation to arise. Heterogeneous nucleation

takes place due to fluctuations of density on particles or surfaces, especially on rough

hydrophobic surfaces (Fischer 2001). For de novo formation of bubbles or for very small

nucleation sites supersaturation pressures exceeding 100 bars (PSS > 100 bars) and very

high SRR must be achieved to overcome the surface tension of pre-existing gas nuclei

(Papadopoulou et al. 2013, Jones et al. 1999). The supersaturation pressure, also called

the “critical pressure” or “initial pressure”, is the maximum applicable pressure Pmax

before bubble formation is observed after decompression. The surface tension γ in relation

to the radius of curvature r of the bubble and the difference in the (partial) pressure inside

and outside the bubble Δp is the critical factor for a bubble to be stable, grow or dissolve


48

(Harvey et al. 1944). This correlation is described by the Young-Laplace-Equation

(Equation 2-5). The minimum radius necessary for a bubble to be stable is called the

critical radius rcrit.

Equation 2-5

;<;!= = 2>∆�

The growth and also the dissolution of a bubble in a polymer is thus dependent on Δp, on

the solubility and diffusion coefficient of the gas in the matrix and on the diffusion of the

gas across the surface surrounding the bubble. (Flook 2000, Harvey et al. 1944, Lubetkin

1994). At very high applied pressures (e.g. 1,000 bars) the pressure overcomes the surface

tension and micronuclei redissolve in the matrix (Harvey et al. 1944). This could imply

that in the high pressure processing of food at pressures up to 6,000 bars most of the gas

nuclei would be dissolved and bubble formation would mainly originate from

spontaneous formation. In addition, additives in polymers and crystallites may serve as

nucleation points.

Cavitation occurs in gas-saturated fluids after a sudden pressure drop, as observed in a

narrow pipe when fluid velocities are high enough (Harvey 1945). At higher pressures,

cavitation effects, namely degassing cavitation, may emerge by desorption of the

dissolved gas (Papadopoulou et al. 2013). In polymers, cavitation generally arises in the

free amorphous phase (Baudet et al. 2009). A cavitation effect of H2 and CO2 in PVDF

after decompression was reported by Boyer et al. (2013).

The phenomenon of tribonucleation arises when two solid surfaces immersed in a liquid

are separated rapidly (Hayward 1967, Ikels 1970). It may be of relevance for HPP when

delamination of two films in a multilayer packaging film occurs due to the difference in

their mechanical properties and compressibility. Reports on the delamination of HPP

packaging can be found in the studies of Bull et al. (2010), Fairclough and Conti (2009),

Fleckenstein et al. (2014), Fraldi et al. (2014), Galotto et al. (2008) and Richter et al.

(2010).

Mathematical models for bubble formation in elastomers, rubbers and other polymers

have been put forward by Baudet et al. (2011) and Kane-Diallo et al. (2016). A good


49

overview of bubble formation during hyperbaric decompression was published by

Papadopoulou et al. (2013).

2.4.2 Factors affecting the formation of bubbles in polymers

Several factors have been reported to be of importance for the phenomenon of explosive

decompression failure (XDF) including material properties (e.g. rigidity and tensile

strength), experimental conditions (e.g. temperature and decompression rate), nature of

the gas and solution properties (Dewimille et al. 1993, Koga et al. 2011, Van Liew and

Burkard 1993). In order to explain XDF, the number of the arising bubbles/blisters per

volume of the matrix (also called the bubble density BD), the maximum size of the

bubbles (BS) and the growth rate (GR) are important. A correlation between BS and BD

was found by Ramesh et al. (1991). Gent and Tompkins (1969) found a positive

correlation between the BD and the GR, whereas Handa and Zhang (2000) observed a

negative correlation.

2.4.2.1 Influence of the nature of the gas and the transport properties on bubble

formation

The solubility of a gas in a polymer influences bubble formation after decompression.

According to Flook (2000) and Ikels (1970), a higher gas solubility leads to a higher

bubble size. This findings are confirmed by Gent and Tompkins (1969), who found

smaller bubbles formed from Ar due to its low solubility in the matrix. Ramesh et al.

(1991) observed that GR is faster with N2 than with CO2. Harvey (1945) had predicted

that bubbles of CO2 would be bigger than those of N2. Sheridan et al. (2000) showed that

CO2 has a greater ability to form bubbles in poly L-lactic acide (PLLA), polyglycolide

(PGA) and their copolymers PLG compared to N2 and He (~23°C, 60 bars). Dipolar

interactions between CO2 and the polar carbonyl groups in the polymer were proposed as

possible explanations by Kazarian et al. (1996) and Bos et al. (1999). Experiments on

cavitation in polymers demonstrated that the damage in PVDF is much more severe with

CO2 than with H2 (Boyer et al. 2013).

Higher diffusion coefficients have been reported to reduce decompression failure in

rubber (Koga et al. 2011, Koga et al. 2013) because the gas is able to escape from the

polymer faster than gases that have a low diffusion coefficient (0< T< 100°C and P< 900

bars). For example, blistering is less pronounced with He because of its higher diffusion


50

rate (2 to 10 times) when compared to H2 or N2 (Koga et al. 2013). In contrast, Chen et

al. (2006) reported a model showing that high diffusion coefficients may strongly increase

the GR in a PS-CO2 system, as the gas is able to enter the bubble more rapidly. These

findings were confirmed by experimental data on the increased blistering effect of CO2

in different thermoplastic polymers when compared to CH4. The effect was explained by

the higher diffusion coefficient of CO2 (Dewimille et al. 1993). The competing effects of

rapid gas entry into bubbles and rapid gas escape from the polymer (high diffusion

coefficient) could not be evaluated using the current literature data.

2.4.2.2 Influence of the material properties on bubble formation

Koga et al. (2011) found that the diffusion coefficient has a higher impact on bubble

formation than the tensile strength of the material. However, both a higher diffusion

coefficient and a higher tensile strength may reduce the formation of bubbles. That is why

Koga et al. (2011) concluded that increasing temperatures (0 to 100°C) have a negative

influence on the blistering behaviour due to the accompanying decrease in tensile

strength. Barth et al. (2013) collected data on the tensile properties of semi-crystalline

thermoplastics under hydrostatic pressure and observed that the tensile strength of

polymers increases with the applied pressure, particularly at high pressures up to 8,000

bars. Considering the prediction by Koga et al. (2011), this would mean that higher

pressures would reduce bubble formation due to an increased tensile strength, a fact that

conflicts with other findings (see Table 2-6). Gent and Tompkins (1969) predicted an

influence of the shear modulus on the crack resistance. They found that higher shear

moduli enable higher supersaturation pressures PSS before blisters arise after

decompression. Barth et al. (2003) reported a linear increase of shear moduli with

pressure. It was reported by Goldman (2009, 2010) that in elastic materials small bubbles

are more likely to form than large bubbles.

In a study on several polymers (PP, PVDF, fluorinated copolymer, PA11, PE-HD) tested

at 40 to 140°C and 100 to 1,000 bars, it was concluded that thermoplastic materials suffer

significantly less from blistering than elastomers (Dewimille et al. 1993 and Jarrin et al.

1994). Also, studies by Sheridan et al. (2000) demonstrated that crystalline polymers (e.g.

PLLA, PGA) withstand the formation of blisters better then amorphous polymers (e.g.

PLG). For plasticised polymers it was reported that pressure and temperature must be

decreased to avoid bubble formation. These results are consistent with the fact that the


51

solubility of gases and the accompanying swelling are higher in elastomers and

amorphous polymers than in thermoplastics and crystalline materials. A fluorinated

copolymer was shown to have the best resistance to bubble formation by CH4 (T< 140°C

and P< 1,000 bars). Fluorinated polymers such as PVDF seem to be unaffected at

temperatures ranging from 70 to 115°C and at pressures up to1,000 bars, while PP had

deformation stability only up to 40°C and 1,000 bars (Dewimille et al. 1993). The results

of Jarrin et al. (1994) showed that PP is only able to withstand blistering below 200 bars

at temperatures up to 100°C. In general, the resistance of PA and fluoride polymers to

XDF was better than polyolefin polymers. Even the type of polyolefin was an important

factor in bubble formation. These results agree with the findings of Grolier and Boyer

(2007) who found a higher solubility for CO2 in PE-MD than in PVDF. Arieli and

Marmur (2011, 2013) found out that BD increased with increasing hydrophobicity of the

material. Similar results had already been observed by Harvey (1945) who recommended

avoiding hydrophobic materials. Harvey reported that bubbles can form spontaneously,

even on hydrophilic surfaces, if the supersaturation pressure PSS and the pressure drop ΔP

are high enough.

Heterogeneous polymers and materials with fillers or other additives have an influence

on the solubility of gases under pressure and therefore may also influence the formation

of bubbles (see section 2.3.1.4). In the work of Harris et al. (1998) sodium chloride

particles were used to alter the BS in PLG. The addition of high concentrations of

surfactants by Gaskins et al. (2001) to a gelatine-water matrix was able to reduce the

formation of bubbles. The work of Yamabe and Nishimura (2010) reveals that Carbon

Black particles added to rubbers enables those materials to be exposed to higher critical

pressures than pure rubbers before bubbles arise.

Ramesh et al. (1991) found that higher molecular weight decreased BS with concurrent

higher BD in a PS polymer (115< T< 125°C and PSS (N2) = 100 bars). On the contrary,

higher molecular weight resulted in a lower BD in PE with CH4 (T = 70°C; Pmax= 100

bars) in the work of Jarrin et al. (1994). Sheridan et al. (2000) observed that increasing

the PLG molecular weight decreases the overall porosity in the matrix for a foaming

process with CO2 (T = 23°C; P = 59 bars). The porosity was calculated from the measured

volume and weight of the matrix compared to the initial density. The authors postulated


52

that the longer polymer chains of high molecular weight polymers may help to withstand

the expansion.

2.4.2.3 Influence of pressurising conditions

The time-pressure profile influences the formation of blisters (Arieli and Marmur 2013,

Dewimille et al. 1993, Van Liew and Burkard 1993). The process involved can be split

into a compression, a pressure holding phase and a decompression phase. In addition, the

maximum applied pressure Pmax and the process temperature are of importance. In some

scientific work the maximum pressure drop ΔP is reported to indicate the maximum

pressure difference that can be applied before bubbles are formed during decompression.

Flook (2000) reported that He tends to require the highest pressure drop for bubbles to

form in human tissue, compared to N2 and Ar. In general, the lower the gas concentration

in the liquid or in the polymer, the higher the required pressure drop for bubbles to form

(Ikels 1970).

Many authors have reported that damage is more severe when a higher Pmax is applied

(Arieli and Marmur 2013, Dewimille et al. 1993, Flook 2000, Jaravel et al. 2011, Yamabe

and Nishimura 2010). This statement was refined by Ramesh et al. (1991), who assumed

that BD increases in PS, whereas BS decreases with increasing applied gas pressure (T =

115°C and 100< P(N2)< 140 bars). Dewimille et al. (1993) found that with increasing

maximum pressure (Pmax< 1,000 bars) the temperature has to be simultaneously reduced

to avoid decompression failure with CH4 in different polymers. Other authors have also

found that lower temperatures result in lower BD (Koga et al. 2011, Van Liew and

Burkard 1993, Gent and Tompkins 1969, Handa and Zhang 2000). At temperatures above

110°C the mean N2 bubble radius in PS was found to increase with temperature (Ramesh

et al. 1991).

The rate of compression was presumed to be of no importance for XDF in the studies of

Jaravel et al. (2011). However, Briscoe et al. (1994) reported that a fast compression rate

is able to reduce decompression failure. Our unpublished work has confirmed that fast

compression rates reduce the amount of XDF. A reason for this may be found in Equation

2-5 which indicates that the critical bubble radius decreases when the difference in the

partial gas pressure inside and outside a bubble is increased. Therefore fast compression

rates lead to a collapse of the pre-existing gas cavities, whereas slow compression enables


53

the diffusion of gas to the inside of the bubble and concomitant compensation of the

pressure difference.

One of the most important ways to reduce XDF is to slow down the decompression rate

(Arieli et al. 2007, Arieli and Marmur 2013, Fairclough and Conti 2009, Flook 2000,

Goldman 2010, Jones et al. 1999, Jaravel et al. 2011, Richter 2014, Van Liew and Burkard

1993). A higher pressure drop and decompression velocity increase the thermodynamic

imbalance in the gas-polymer system. The resulting triaxial tension may induce cavitation

or tribonucleation via the mechanical separation of polymer films (e.g. delamination).

The model of Baudet et al. (2011) for the prediction of blistering in semi-crystalline

polymers indicates that higher decompression rates increases the amount of voids. The

size of the damage is higher in the centre of the polymer due to a concentration gradient

of the gas, especially at enhanced decompression rates. It has to be taken into account in

any case that if the decompression rate is faster than the gas is able to escape from the

polymer then bubbles will form spontaneously or the gas will diffuse into pre-existing

micronuclei to overcome the supersaturation state.

Table 2-6 summarises the results of the literature review on XDF, listing the nature and

transport properties of the gases, the material properties and process conditions. The

systems studied in the scientific work are summarised in Table 2-7.

Table 2-6 Parameters influencing bubble formation (BF), bubble growth rate (GR), bubble size (BS) and the number of bubbles expressed as bubble density (BD).

Parameter Observations Reference Nature of the gas and transport properties of the gas in the matrix

Solubility ↑ ↑ BF (Flook 2000; Jarrin et al. 1994)

Solubility ↑ ↑ GR (Harvey et al. 1944)

Solubility ↑ ↑ BS (Ramesh et al. 1991)

Solubility ↑ ↓ BS ↑ BD (Handa and Zhang 2000)

Solubility ↑ BS(CO2) > BS(N2) > BS(Ar) (Gent and Tompkins 1969)

Solubility BF(Ar) > BF(N2) > BF(He) (Flook 2000)

Solubility BF(CO2) >> BF(N2) ≈ BF(He) (Sheridan et al. 2000)

Gas nature GR(CO2)> GR(N2) (Ramesh et al. 1991)

Gas nature BS(CO2)> BS(N2) (Harvey 1945)

Gas nature BF(CO2) > BF(O2) (Götz and Weisser 2002)

Gas nature BF(CO2) > BF(H2S) > BF(CH4) (Dewimille et al. 1993) Diffusion coefficient ↑

BF(N2) > BF(H2) > BF(He) (Koga et al. 2013)

↓ BS


54

Desorption diffusion ↑

↓ BF (Koga et al. 2011, Koga et al. 2013)

Diffusion coefficient ↑

↓ BF (Jarrin et al. 1994, Koga et al. 2011)

Diffusion coefficient ↑

↑ BS ↑ BD (Chen et al. 2006)

Material properties

Molecular weight ↑ ↓ GR and BS ↑ BD (Ramesh et al. 1991)

Molecular weight ↑ ↓ BD (Jarrin et al. 1994)

Molecular weight ↑ ↓ BF (Sheridan et al. 2000)

Shear modulus ↑ Maximum pressure drop increases before bubbles arise

(Gent and Tompkins 1969)

Mechanical or tensile strength of the material ↑

↓ BF (Yamabe and Nishimura 2010, Koga et al. 2011, Jarrin et al. 1994)

Viscosity of the liquid ↑

ΔP hast to be decreased to avoid BF

(Ikels 1970)

Elasticity ↑ (e.g. Elastomers ↔ Thermoplastics

↑ BF(Thermoplastics) < BF(Elastomers) (Dewimille et al. 1993, Jarrin et al. 1994)

Elasticity ↑ ↓ BS (Goldman 2009, 2010) Plasticisation of polymer by gas ↑

↓ GR (Chen et al. 2006)

Crystallinity BF(crystalline polymer) < BF(amorphous

polymer) (Sheridan et al. 2000, Jarrin et al. 1994)

Rigidity ↑ ↓ BF (Richter et al. 2010)

Hydrophobicity ↑ ↑ BF (Arieli and Marmur 2011, 2013, Harvey 1945)

Surface tension ↑ ↓ BS (Goldman 2009)

Surface tension ↑ ↑ BF (Gaskins et al. 2001)

Surface tension ↑ ↑ BD (Van Liew and Burkard 1993)

Homogeneity of the material ↑

↓ BF (Koga et al. 2013, Jarrin et al. 1994)

Nature of polymer

BF (PVDF) < BF (Teflon) < BF(PP) BF (PVDF) < BF (PA11) < BF (PE-HD) < BF (PP) (at T< 80°C)

(Dewimille et al. 1993, Jarrin et al. 1994)

Experimental conditions Pressure-time profile

Compression rate ↑ ↓ BD (Briscoe et al. 1994) Decompression rate ↑

↑ BS ↑ BD (Baudet et al. 2011)

Decompression rate ↑

↑ BS ↓ BD (Arora et al. 1998)


55


↑ BD (Arieli et al. 2007)


↑ BF

(Jaravel et al. 2011, Koga et al. 2011, Van Liew and Burkard 1993, Jones et al. 1999, Richter 2011)

Maximum applied pressure Pmax; ↑ Maximum pressure drop ΔP ↑

↑ BF

(Yamabe and Nishimura 2010, Koga et al. 2011, Dewimille et al. 1993, Arieli and Marmur 2013)

Maximum pressure drop ΔP ↑

↑ GR ↑ BD (Van Liew and Burkard 1993)

Max. applied pressure ↑

↓ BS ↑ BD (Gent and Tompkins 1969, Ramesh et al. 1991, Arieli et al. 2007, Arora et al. 1998)

Pressure holding time ↑

↑ BF (Masuda et al. 1992, Koga et al. 2011, Koga et al. 2013)

Pressure holding time

No effect (Fradin et al. 1998)

Temperature

Temperature ↑ (0 – 140°C)

↑ BF

(Dewimille et al. 1993, Koga et al. 2011, Van Liew and Burkard 1993, Richter et al. 2010)


↑ GR and BS ↓ BD (Arora et al. 1998)


↑ GR and BS (Ramesh et al. 1991)


↓ BS ↑ BD (Handa and Zhang 2000)


↑ BD (Gent and Tompkins 1969)


↓ BD (Handa and Zhang 2000)

Combined effect of T ↑ and P ↑

↑ BF (Dewimille et al. 1993)

Table 2-7 Overview of systems studied in the literature and the experimental conditions

Gas Polymer / Matrix Applied temperature [°C]

Applied pressure [bar]

Reference

Static gas pressure (two-phase system) CO2 PS 40 – 120 70 – 480 (Arora et al. 1998) CH4, Fluorinated polymer, 40 – 140 100 – (Dewimille et al.


56

CO2, H2S

fluorinated copolymer, PP

1,000 1993)

CO2 PMMA 10 – 95 34 (Handa and Zhang 2000)

Ar, N2, CO2

Butadiene-styrene copolymer, natural rubber (pale crepe)

10 – 65 ~50 (Gent and Tompkins 1969)

CO2, CH4, H2S

PVDF, PA11, PE-HD, PP

150 1,000 (Jarrin et al. 1994)

CO2, He, N2

PLG, PGA, PLLA 25 59 (Sheridan et al. 2000)

N2, CO2 PS 105 - 135 55 - 140 (Ramesh et al. 1991)

Dynamic gas pressure; Pressure difference between upstream and downstream (two-phase system)

H2 EPDM, VMQ, HNBR

0 - 100 950 (Koga et al. 2011)

H2, He, N2

EPDM, VMQ 25 100 (Koga et al. 2013)

CO2 PS model (Chen et al. 2006)

H2 EDPM, NBR containing carbon black or silica fillers

30 100 (Yamabe and Nishimura 2010)

CO2 PVDF Theoretical considerations + experiments

250 (Baudet et al. 2011)

Elastomers Literature research

(Briscoe et al. 1994)

H2 Vinyltrimetoxy-silane

~20 270 (Jaravel et al. 2011)

Isostatic pressure (three-phase system) Ar, N2, He

Human tissue unknown 2 – 6 (Flook 2000)

O2 PE film in water 20 – 30 2,000 bar (Sterr et al. 2015b)

N2, CO2 MAP made from PE, PET, PA, EVOH

5, 23, 40 6,000 bar (Sterr et al. 2015a)

Air in water

Silicon wafer 25 10 (Arieli and Marmur 2011)

Air, O2 Prawn 25 2-8 (Arieli et al. 2007) Air in water

Silicon wafer 25 1.5 – 4 (Arieli and Marmur 2013)

Air, O2 Animal tissue (Harvey 1945)

CO2, O2 PA/PE bag ~23 5,000 (Götz and Weisser 2002)

He, Ar, N2

Metal ball in glass tube filled with

25 Pressure drop

(Ikels 1970)


57

liquid (olive oil, glycerol-water)

ΔP < 10 bars

Air Agar gel unknown unknown (Gaskins et al. 2001)

N2, CO2, O2

Packaging made from PE, PET, PA, EVOH

10 – 70 6,000 (Richter 2011)

- Packaging made from PE, PET, EVOH, PP, PA etc.

20 – 40 6,000 (Masuda et al. 1992)

Air Packaging made from PE, PA, PET etc. filled with water

25 2,000 (Fradin et al. 1998)

Air Packaging made from PP

70 6,950 (Fairclough and Conti 2009)

Theoretical considerations

Bubble growth in soft elastic materials


(Goldman 2009)

N2 Stability of bubbles in soft elastic materials


(Goldman 2010)

Gases Liquid Literature research

(Jones et al. 1999)

Gases Animal tissue, liquids

Theoretical considerations, literature research

(Harvey et al. 1944)

Air, N2 Blood, human tissue Mathematical simulations

2 (Van Liew and Burkard 1993)

2.4.3 Bubble formation in packaging used for HPP

Insufficient data is available on the formation of bubbles in multilayer films as part as

modified atmosphere packaging used for foods processed at high pressure (4,000 to 6,000

bars) and moderate temperatures. However, some results and considerations were gleaned

from the literature. Increasing the headspace volume and/or high internal gas pressure pi

(pi>patm) in packaging used for HPP lead to increased blistering (Fairclough and Conti

2009, Sterr et al. 2015a). A reason might be that the enhanced internal gas pressure

increases the amount of gas dissolved in the material even under atmospheric conditions.

Also, the packaging geometry might play a role and this has been considered by a few

authors (Fairclough and Conti 2009, Sterr et al. 2015a, Richter 2011). The edges and folds

of tray packaging seem to be more vulnerable to blistering. Additionally, it has to be taken

into account that gases do not merely dissolve in the polymeric part of the packaging but


58

also in the food product and may therefore also form bubbles there (Al-Nehlawi et al.

2014). To our knowledge, no scientific work has dealt with the interaction and combined

effect of polymer and packed food. Also, the adiabatic heating of the gas, food and

polymers, and even the pressure transmitting fluid, may influence the amount of

blistering. In general, the effect of the pressure transmitting medium is unknown (Barth

et al. 2013). An interesting fact mentioned by Muth et al. (2001) was that visible blisters

and foam like structures can be reduced by thermal treatment above the glass transition

temperature of the material for a specific time. In their studies, opaque PVC samples

became transparent again after treatment at 75°C for several hours.

2.5 Conclusion

Even though most of the studies about the solution, diffusion and permeation of gases in

polymers under pressure were performed with significantly different measurement

methods under differing temperature and pressure conditions, same common conclusions

can be drawn. Most authors suggest that under the low temperature conditions used for

HPP (below 50°C) gas solubility decreases with increasing temperature and that it

increases with increasing pressure at medium pressures. However, a possible saturation

behaviour was postulated at higher pressures by a few authors no exact predictions can

be made for pressures up to 6,000 bars. The data suggest that CO2 has a higher solubility

in all the tested polymers than gases that have low critical temperatures such as O2, N2

and He. Although numerous studies have been performed on the sorption of CO2 in

common polymers, little information is available about the solubility of O2 at high

pressures and low temperatures. A lower density or a higher content of amorphous phase

in polymers results in an increased solubility of gases and therefore swelling and

plasticisation may change the amount of gas that is able to dissolve in the polymer. Only

a few authors have taken swelling into account when determining gas solubility. It is

generally assumed that the relative volume change increases with pressure and

temperature at pressures below 100 bars in the low temperature range. However, the

influence of the competing effect of plasticisation of the polymer chains and the

compression of the free volume on the transport properties of gases in polymers is still

difficult to predict, especially at very high pressures. The glass transition temperature is

expected to decrease with applied pressure, but some authors have observed that Tg


59

increases at pressures above 1,000 bars. Filler particles have diverse effects on the

transport properties, depending on the affinity of the particles to the diffusing gas

molecules. The sorption and desorption behaviour shows strong hysteresis. Permeation

increases with pressure for highly soluble gases such as CO2 and decreases with

increasing pressure for sparingly soluble gases such as N2 and O2. Permeability

coefficients are reduced 2 fold at pressures up to 100 bars, but are highly reduced (70

fold) at pressures up to 2,000 bars.

Based on the available data it is not possible to make reliable predictions about the

transport properties of gases under the pressure conditions used for the high pressure

processing of food. No relevant data are available about the transport properties of

supercritical fluids. Further research is necessary to measure the solubility, diffusion and

permeation of relevant gases (O2, N2, CO2) in packaging-related polymers (PE, PET, PA,

PP) at isostatic pressures up to 6,000 bars and at low temperatures. The measurement of

CO2 permeation at high hydrostatic pressures could be possible by adapting the

measurement setup described by Sterr et al. (2015b). The competing effects of swelling,

plasticising and compression should be taken into account.

Concerning the influence of certain parameters on the formation of bubbles some

conclusions can be drawn. . However, the literature contained conflicting data.

Differences in the measurement methods and processing conditions may be the reason for

the conflicting results. It should be pointed out that in some of the published work

information on important parameters is missing. Opinion about which type of gas leads

to the most severe decompression failure differs in the literature. However, most authors

agree that a higher solubility enhances blistering. Bubble formation with He is

comparably rare, whereas the use of CO2 affects the materials more markedly. It is not

clear yet how the diffusion coefficient influences packaging failure as a high diffusion

coefficient allows both high access of the gas to the bubbles in the compression phase and

high desorption of gas out of the polymer in the decompression phase. Material properties

such as crystallinity, tensile strength, shear moduli, surface energy, viscosity, molecular

weight and hydrophobicity affect the polymer behaviour during decompression. In

general, thermoplastic polymers are more stable than elastomers and fluorinated

polymers. Also, polyamides show better performance than polyolefin polymers.

Considering these facts, a polymer with a higher crystalline content such as high density


60

polyethylene may have improved packaging performance than low density polyethylene

for use as the inner layer of packaging for HPP. In multilayer films used in the food

industry, differing mechanical properties and solution effects of the single layers may

lead to increased dissolution of gas at interstitial sites. Delamination may also arise.

Inhomogeneities acting as nucleation points in polymers increase XDF. However, filler

particles have the ability to control the uptake of gas and hence also the formation of

bubbles. Apart from the solubility of the gas in the polymers, the decompression rate

seems to be the most important factor. Most authors report that a slow decompression rate

results in reduced blistering. Increasing the temperature seems to promote bubble

formation. However, no adiabatic heating or heat transfer from the pressure transmitting

fluid in more complex systems, such as those used in HPP, was taken into account in any

of the relevant literature. Other parameters such as a faster compression rate and a lower

maximum applied pressure and lower pressure drop also aid crack prevention. In

summary, the occurrence and behaviour of explosive decompression failure is not easy to

predict due to the many parameters involved and the fact no relevant scientific studies

have been performed at pressures up to 6,000 bars on sealed packaging. Future studies on

bubble formation in modified atmosphere packaging for high pressure processing should

focus on the influence of the food products and their adiabatic heating and heat transfer

properties. The transport, mechanical and structural properties of different types of

polymers should be taken into account.


61

2.6 Publication bibliography

AL-NEHLAWI, A.; GURI, S.; GUAMIS, B.; SALDO, J. (2014): Synergistic effect of carbon

dioxide atmospheres and high hydrostatic pressure to reduce spoilage bacteria on

poultry sausages. In LWT-Food Science and Technology 58 (2), pp. 404–411.

AREERAT, S.; HAYATA, Y.; KATSUMOTO, R.; KEGASAWA, T.; EGAMI, H.; OHSHIMA, M.

(2002): Solubility of carbon dioxide in polyethylene/titanium dioxide composite

under high pressure and temperature. In J. Appl. Polym. Sci. 86 (2), pp. 282–288.

DOI: 10.1002/app.10944.

ARIELI, R.; MARMUR, A. (2011): Decompression sickness bubbles: Are gas micronuclei

formed on a flat hydrophobic surface? In Respiratory Physiology & Neurobiology

177 (1), pp. 19–23. DOI: 10.1016/j.resp.2011.02.013.

ARIELI, R.; MARMUR, A. (2013): Dynamics of gas micronuclei formed on a flat

hydrophobic surface, the predecessors of decompression bubbles. In Respiratory

Physiology & Neurobiology 185 (3), pp. 647–652. DOI: 10.1016/j.resp.2012.11.020.

ARIELI, Y.; KATSENELSON, K.; ARIELI, R. (2007): Bubble reduction after decompression

in the prawn Palaemon elegans by pretreatment with hyperbaric oxygen. In Undersea

and Hyperbaric Medical Society 34 (5), pp. 369–378.

ARORA, K. A.; LESSER, A. J.; MCCARTHY, T. J. (1998): Preparation and

characterization of microcellular polystyrene foams processed in supercritical carbon

dioxide. In Macromolecules 31 (14), pp. 4614–4620. DOI: 10.1021/ma971811z.

ATKINS, P. W.; PAULA, J. de (2006): Physikalische Chemie. 4., vollst. überarb. Aufl.

Weinheim [u.a.]: VCH.

BARRER, R. M. (1951): Diffusion in and through solids. Cambridge: University Press

(The Cambridge series of physical chemistry).

BARTH, R. R.; SIMMONS, K. L.; SAN MARCHI, C. (2013): Polymers for hydrogen

infrastructure and vehicle fuel systems. Applications, properties, and gap analysis.

Edited by Sandia National Laboratories. Sandia Corporation. Albuquerque, New

Mexico, U.S.A.


62

BATTINO, R.; CLEVER, H. L. (1966): The solubility of gases in liquids. In Chem. Rev. 66

(4), pp. 395–463. DOI: 10.1021/cr60242a003.

BAUDET, C.; GRANDIDIER, J.-C.; CANGÉMI, L. (2009): A two-phase model for the

diffuso-mechanical behaviour of semicrystalline polymers in gaseous environment.

In International Journal of Solids and Structures 46 (6), pp. 1389–1401. DOI:

10.1016/j.ijsolstr.2008.11.010.

BAUDET, C.; GRANDIDIER, J.-C.; CANGÉMI, L. (2011): A damage model for the

blistering of polyvinylidene fluoride subjected to carbon dioxide decompression. In

Journal of the Mechanics and Physics of Solids 59 (9), pp. 1909–1926. DOI:

10.1016/j.jmps.2011.04.010.

BERENS, A. R.; HUVARD, G. S.; KORSMEYER, R. W.; KUNIG, F. W. (1992): Application

of compressed carbon dioxide in the incorporation of additives into polymers. In J.

Appl. Polym. Sci. 46 (2), pp. 231–242. DOI: 10.1002/app.1992.070460204.

BONAVOGLIA, B.; STORTI, G.; MORBIDELLI, M.; RAJENDRAN, A.; MAZZOTTI, M. (2006):

Sorption and swelling of semicrystalline polymers in supercritical CO2. In Journal of

Polymer Science Part B: Polymer Physics 44 (11), pp. 1531–1546.

BOS, A.; PÜNT, I. G. M.; WESSLING, M.; STRATHMANN, H. (1999): CO2-induced

plasticization phenomena in glassy polymers. In Journal of Membrane Science 155

(1), pp. 67–78. DOI: 10.1016/S0376-7388(98)00299-3.

BOYER, S. A. E.; GERLAND, M.; CASTAGNET, S. (2013): Gas environment effect on

cavitation damage in stretched polyvinylidene fluoride. In Polym Eng Sci 54 (9),

pp. 2139–2146. DOI: 10.1002/pen.23759.

BOYER, S. A. E.; GROLIER, J.-P. E. (2005a): Modification of the glass transitions of

polymers by high-pressure gas solubility. In Pure and applied chemistry 77 (3),

pp. 593–603.

BOYER, S. A. E.; GROLIER, J.-P. E. (2005b): Simultaneous measurement of the

concentration of a supercritical gas absorbed in a polymer and of the concomitant

change in volume of the polymer. The coupled VW-pVT technique revisited. In

Polymer 46 (11), pp. 3737–3747.


63

BOYER, S. A. E.; GROLIER, J.-P. E.; YOSHIDA, H.; HAUDIN, J.-M.; CHENOT, J.-L.

(2011): Thermodynamics and thermokinetics to model phase transitions of polymers

over extended temperature and pressure ranges under various hydrostatic fluids. In

Thermodynamics-Interaction Studies-Solids, Liquids and Gases.

BOYER, S. A. E.; KLOPFFER, M.-H.; MARTIN, J.; GROLIER, J.-P. E. (2007): Supercritical

gas–polymer interactions with applications in the petroleum industry. Determination

of thermophysical properties. In J. Appl. Polym. Sci. 103 (3), pp. 1706–1722. DOI:

10.1002/app.25085.

BOYER, S. A. E.; RANDZIO, S. L.; GROLIER, J.-P. E. (2006): Thermal expansion of

polymers submitted to supercritical CO2 as a function of pressure. In Journal of

Polymer Science Part B: Polymer Physics 44 (1), pp. 185–194. DOI:

10.1002/polb.20674.

BRANDRUP, J.; IMMERGUT, E. H.; GRULKE, E. A. (Eds.) (1999): Polymer handbook.

Volume 1 & 2. 4.th ed. Hoboken, N. J.: Wiley.

BRISCOE, B. J.; KELLY, C. T. (1996): The effect of structure on gas solubility and gas

induced dilation in a series of poly(urethane) elastomers. In Polymer 37 (15),

pp. 3405–3410. DOI: 10.1016/0032-3861(96)88488-6.

BRISCOE, B. J.; MAHGEREFTEH, H. (1984): A novel technique for the quantitative

measurement of gaseous uptake in organic polymers at high pressures. In Journal of

Physics E: Scientific Instruments 17 (6), p. 483.

BRISCOE, B. J.; SAVVAS, T.; KELLY, C. T. (1994): “Explosive decompression failure” of

rubbers: A review of the origins of pneumatic stress induced rupture in elastomers. In

Rubber Chemistry and Technology 67 (3), pp. 384–416. DOI: 10.5254/1.3538683.

BRISCOE, B. J.; ZAKARIA, S. (1990): Polymer-gas interactions at high pressure: the use

of ultrasonic probes. In Polymer 31 (3), pp. 440–447.

BRISCOE, B. J.; ZAKARIA, S. (1991): Interaction of CO2 gas with silicone elastomer at

high ambient pressures. In J. Polym. Sci. B Polym. Phys. 29 (8), pp. 989–999. DOI:

10.1002/polb.1991.090290809.


64

BRISCOE, B. J.; ZAKARIA, S. (1992): Sorption and dilation of silicone elastomer

composites at high gas pressures: The role of interfacial quality. In J. Polym. Sci. B

Polym. Phys. 30 (9), pp. 959–969. DOI: 10.1002/polb.1992.090300904.

BUDD, P. M.; MCKEOWN, N. B.; FRITSCH, D. (2005): Free volume and intrinsic

microporosity in polymers. In J. Mater. Chem. 15 (20), pp. 1977–1986. DOI:

10.1039/B417402J.

BULL, M. K.; STEELE, R. J.; KELLY, M.; OLIVIER, S. A.; CHAPMAN, B. (2010):

Packaging under pressure: Effects of high pressure, high temperature processing on

the barrier properties of commonly available packaging materials. In Innovative

Food Science & Emerging Technologies 11 (4), pp. 533–537.

CAMPBELL, J. (1968): The tribonucleation of bubbles. In Journal of Physics D: Applied

Physics 1 (8), p. 1085.

CAMPION, R. P.; MORGAN, G. J. (1992): High pressure permeation and diffusion of

gases in polymers of different structures. In Plastics, Rubber and Composites

Processing and Applications 17 (1), pp. 51–58.

CHANG, S.-H.; PARK, S.-C.; SHIM, J.-J. (1998): Phase equilibria of supercritical fluid–

polymer systems. In The Journal of Supercritical Fluids 13 (1-3), pp. 113–119. DOI:

10.1016/S0896-8446(98)00042-4.

CHEN, X.; FENG, J. J.; BERTELO, C. A. (2006): Plasticization effects on bubble growth

during polymer foaming. In Polym Eng Sci 46 (1), pp. 97–107. DOI:

10.1002/pen.20434.

CHIOU, J. S.; BARLOW, J. W.; PAUL, D. R. (1985a): Plasticization of glassy polymers by

CO2. In J. Appl. Polym. Sci. 30 (6), pp. 2633–2642. DOI:

10.1002/app.1985.070300626.

CHIOU, J. S.; BARLOW, J. W.; PAUL, D. R. (1985b): Polymer crystallization induced by

sorption of CO2 gas. In Journal of Applied Polymer Science 30 (9), pp. 3911–3924.

DOI: 10.1002/app.1985.070300929.

CHOW, T. S. (1980): Molecular interpretation of the glass transition temperature of

polymer-diluent systems. In Macromolecules 13 (2), pp. 362–364. DOI:

10.1021/ma60074a029.


65

CORWIN, H.; SHELLHAMMER, T. H. (2002): Combined carbon dioxide and high pressure

inactivation of pectin methylesterase, polyphenol oxidase, Lactobacillus plantarum

and Escherichia coli. In Journal of Food Science 67 (2), pp. 697–701.

CRANK, J. (1975): The mathematics of diffusion. 2nd ed. London, UK: Clarendon press

Oxford (3).

DEWIMILLE, B.; MARTIN, J.; JARRIN, J. (1993): Behaviour of thermoplastic polymers

during explosive decompressions in a petroleum environment. In Le Journal de

Physique IV 3 (C7), p. 1559.

DUNCAN, B.; URQUHART, J.; ROBERTS, S. (January / 2005): Review of measurement

and modelling of permeation and diffusion in polymers. NPL Report DEPC MPR

012. Edited by National Physical Laboratory. Middlesex.

ENNS, T.; SCHOLANDER, P. F.; BRADSTREET, E. D. (1965): Effect of hydrostatic

pressure on gases dissolved in water. In J. Phys. Chem. 69 (2), pp. 389–391. DOI:

10.1021/j100886a005.

FAIRCLOUGH, J. P. A.; CONTI, M. (2009): Influence of ultra-high pressure sterilization

on the structure of polymer films. In Packaging Technology & Science 22 (5),

pp. 303–310. DOI: 10.1002/pts.858.

FISCHER, S. (2001): Blasenbildung von in Flüssigkeiten gelösten Gasen. Technische

Universität München, Universitätsbibliothek.

FLACONNÈCHE, B.; MARTIN, J.; KLOPFFER, M.-H. (2001a): Permeability, diffusion and

solubility of gases in polythylene, polyamide 11 and poly(vinylidene fluoride). In Oil

Gas Sci. Technol. – Rev. IFP Energies nouvelles 56 (3), pp. 261–278.

FLACONNÈCHE, B.; MARTIN, J.; KLOPFFER, M.-H. (2001b): Transport properties of

gases in polymers: Experimental methods. In Oil Gas Sci. Technol. – Rev. IFP

Energies nouvelles 56 (3), pp. 245–259. DOI: 10.2516/ogst:2001022.




10.1002/pts.2018.


66

FLEMING, G. K.; KOROS, W. J. (1986): Dilation of polymers by sorption of carbon

dioxide at elevated pressures. 1. Silicone rubber and unconditioned polycarbonate. In

Macromolecules 19 (8), pp. 2285–2291. DOI: 10.1021/ma00162a030.

FLOOK, V. (2000): The physics and physiology of decompression. In Eur J Underwater

Hyperbaric Med 1 (1), pp. 8–13.




FRALDI, M.; CUTOLO, A.; ESPOSITO, L.; PERRELLA, G.; Carbone, Maria Giovanna

Pastore; SANSONE, L. et al. (2014): Delamination onset and design criteria of

multilayer flexible packaging under high pressure treatments. In Innovative Food

Science & Emerging Technologies. DOI: 10.1016/j.ifset.2014.02.016.

FUJITA, H. (1961): Diffusion in polymer-diluent systems. In : Fortschritte Der

Hochpolymeren-Forschung: Springer, pp. 1–47.

GALOTTO, M. J.; ULLOA, P. A.; HERNÁNDEZ, D.; FERNÁNDEZ-MARTÍN, F.; GAVARA, R.;

GUARDA, A. (2008): Mechanical and thermal behaviour of flexible food packaging

polymeric films materials under high pressure/temperature treatments. In Packaging

Technology & Science 21 (5), pp. 297–308. DOI: 10.1002/pts.807.

GASKINS, N.; VANN, R. D.; HOBBS, G. W.; SWINGLE, M.; LEE, S.; NEEDHAM, D.

(2001): Surface tension and bubble formation in agar gelatin. In Undersea and

Hyperbaric Medical Society 28, pp. 56–59.

GENT, A. N.; TOMPKINS, D. A. (1969): Nucleation and growth of gas bubbles in

elastomers. In Journal of Applied Physics 40 (6), pp. 2520–2525. DOI:

10.1063/1.1658026.

GOLDMAN, S. (2009): Generalizations of the Young–Laplace equation for the pressure

of a mechanically stable gas bubble in a soft elastic material. In J. Chem. Phys 131

(18), p. 184502. DOI: 10.1063/1.3259973.

GOLDMAN, S. (2010): Free energy wells for small gas bubbles in soft deformable

materials. In J. Chem. Phys 132 (16), p. 164509. DOI: 10.1063/1.3394940.


67

GÖTZ, J.; WEISSER, H. (2002): Permeation of aroma compounds through plastic films

under high pressure: in-situ measuring method. In Innovative Food Science &

Emerging Technologies 3 (1), pp. 25–31.

GRANDIDIER, J.-C.; BAUDET, C.; BOYER, S. A.E.; KLOPFFER, M.-H.; CANGÉMI, L.

(2015): Diffuso-kinetics and diffuso-mechanics of carbon dioxide / polyvinylidene

fluoride system under explosive gas decompression. Identification of key diffuso-

elastic couplings by numerical and experimental confrontation. In Oil Gas Sci.

Technol. – Rev. IFP Energies nouvelles 70 (2), pp. 251–266. DOI:

10.2516/ogst/2013198.

GROLIER, J.-P. E.; BOYER, S. A. E. (2007): Solubility of gases in polymers. In T. M.

Letcher (Ed.): Development and applications in solubility. Cambridge: RSC Pub,

pp. 249–260.

HANDA, Y. P.; ZHANG, Z. (2000): A new technique for measuring retrograde

vitrification in polymer-gas systems and for making ultramicrocellular foams from

the retrograde phase. In J. Polym. Sci. B Polym. Phys. 38 (5), pp. 716–725. DOI:

10.1002/(SICI)1099-0488(20000301)38:5<716::AID-POLB9>3.0.CO;2-N.

HARRIS, L. D.; KIM, B.-S.; MOONEY, D. J. (1998): Open pore biodegradable matrices

formed with gas foaming. In Journal of biomedical materials research 42 (3),

pp. 396–402.

HARVEY, E. N. (1945): Decompression sickness and bubble formation in blood and

tissues. In Bulletin of the New York Academy of Medicine 21 (10), p. 505.

HARVEY, E. N.; BARNES, D. K.; MCELROY, W. D.; WHITELEY, A. H.; PEASE, D. C.;

COOPER, K. W. (1944): Bubble formation in animals. I. Physical factors. In J. Cell.

Comp. Physiol. 24 (1), pp. 1–22. DOI: 10.1002/jcp.1030240102.

HAYWARD, A. T. J. (1967): Tribonucleation of bubbles. In Br. J. Appl. Phys. 18 (5),

pp. 641–644. DOI: 10.1088/0508-3443/18/5/314.

HILIC, S.; BOYER, S. A. E.; PÁDUA, A. A. H.; GROLIER, J.-P. E. (2001): Simultaneous

measurement of the solubility of nitrogen and carbon dioxide in polystyrene and of

the associated polymer swelling. In Journal of Polymer Science Part B: Polymer

Physics 39 (17), pp. 2063–2070. DOI: 10.1002/polb.1181.


68

IKELS, K. G. (1970): Production of gas bubbles in fluids by tribonucleation. In Journal

of Applied Physiology 28 (4), pp. 524–527.

JARAVEL, J.; CASTAGNET, S.; GRANDIDIER, J.-C.; BENOÎT, G. (2011): On key

parameters influencing cavitation damage upon fast decompression in a hydrogen

saturated elastomer. In Polymer Testing 30 (8), pp. 811–818. DOI:

10.1016/j.polymertesting.2011.08.003.

JARRIN, J.; DEWIMILLE, B.; DEVAUX, E. (1994): Blistering of thermoplastic materials

used in the petroleum industry. In : SPE Annual Technical Conference and

Exhibition. SPE Annual Technical Conference and Exhibition. New Orleans,

Louisiana, 1994-09-25. Society of Petroleum Engineers. New Orleans, LA, U.S.A.,

pp. 203–214.

JONES, S. F.; EVANS, G. M.; GALVIN, K. P. (1999): Bubble nucleation from gas cavities

— a review. In Advances in Colloid and Interface Science 80 (1), pp. 27–50. DOI:

10.1016/S0001-8686(98)00074-8.

JONES PARRY, E.; TABOR, D. (1973): Effect of hydrostatic pressure on the mechanical

properties of polymers. A brief review of published data. In J Mater Sci 8 (10),

pp. 1510–1516. DOI: 10.1007/BF00551675.

JORDAN, S. M.; KOROS, W. J. (1990): Permeability of pure and mixed gases in silicone

rubber at elevated pressures. In Journal of Polymer Science Part B: Polymer Physics

28 (6), pp. 795–809. DOI: 10.1002/polb.1990.090280602.

KANE-DIALLO, O.; CASTAGNET, S.; NAIT-ALI, A.; BENOÎT, G.; GRANDIDIER, J.-C.

(2016): Time-resolved statistics of cavity fields nucleated in a gas-exposed rubber

under variable decompression conditions – Support to a relevant modeling

framework. In Polymer Testing 51, pp. 122–130. DOI:

10.1016/j.polymertesting.2016.03.004.

KAZARIAN, S. G. (2000): Polymer processing with supercritical fluids. In Polymer

Science 42 (1), pp. 78–101.

KAZARIAN, S. G.; VINCENT, M. F.; BRIGHT, F. V.; LIOTTA, C. L.; ECKERT, C. A. (1996):

Specific intermolecular interaction of carbon dioxide with polymers. In Journal of

the American Chemical Society 118 (7), pp. 1729–1736. DOI: 10.1021/ja950416q.


69

KLOPFFER, M.-H.; FLACONNÈCHE, B. (2001): Transport properties of gases in polymers.

Bibliographic Review. In Oil Gas Sci. Technol. – Rev. IFP Energies nouvelles 56

(3), pp. 223–244.

KLOTZ, I. M. (1963): Variation of solubility with depth in the ocean. A thermodynamic

analysis. In Limnol. Oceangr. 8 (2), pp. 149–151. DOI: 10.4319/lo.1963.8.2.0149.

KOGA, A.; UCHIDA, K.; YAMABE, J.; NISHIMURA, S. (2011): Evaluation on high-

pressure hydrogen decompression failure of rubber O-ring using design of

experiments. In International Journal of Automotive Engineering 2 (4), pp. 123–129.

KOGA, A.; YAMABE, T.; SATO, H.; UCHIDA, K.; NAKAYAMA, J.; YAMABE, J.;

NISHIMURA, S. (2013): A visualizing study of blister initiation behavior by gas

decompression. In Tribology Online 8 (1), pp. 68–75.

KOUTCHMA, T.; SONG, Y.; SETIKAITE, I.; JULIANO, P.; BARBOSA-CÁNOVAS, G. V.;

DUNNE, C. P.; PATAZCA, E. (2009): Packaging evaluation for high-pressure high-

temperature sterilization of shelf-stable foods. In Journal of Food Process

Engineering 33 (6), pp. 1097–1114. DOI: 10.1111/j.1745-4530.2008.00328.x.

KULKARNI, S. S.; STERN, S. A. (1983): The diffusion of CO2, CH4, C2H4, and C3H8 in

polyethylene at elevated pressures. In J. Polym. Sci. Polym. Phys. Ed. 21 (3),

pp. 441–465. DOI: 10.1002/pol.1983.180210310.

LEI, Z.; OHYABU, H.; SATO, Y.; INOMATA, H.; SMITH, R. L. (2007): Solubility, swelling

degree and crystallinity of carbon dioxide–polypropylene system. In The Journal of

Supercritical Fluids 40 (3), pp. 452–461. DOI: 10.1016/j.supflu.2006.07.016.

LEWIS, E.L.V.; DUCKETT, R. A.; WARD, I. M.; FAIRCLOUGH, J. P. A.; RYAN, A. J.

(2003): The barrier properties of polyethylene terephthalate to mixtures of oxygen,

carbon dioxide and nitrogen. In Polymer 44 (5), pp. 1631–1640. DOI:

10.1016/S0032-3861(02)00933-3.

LORGE, O.; BRISCOE, B. J.; DANG, P. (1999): Gas induced damage in poly (vinylidene

fluoride) exposed to decompression. In Polymer 40 (11), pp. 2981–2991.

LUBETKIN, S. D. (1994): Bubble nucleation and growth. In D. J. Wedlock (Ed.):

Controlled particle, droplet and bubble formation. Oxford: Butterworth-Heinemann

Ltd, Oxford, pp. 159–190.


70

LUDWIG, H.; MACDONALD, A. G. (2005): The significance of the activity of dissolved

oxygen, and other gases, enhanced by high hydrostatic pressure. In Comparative

biochemistry and physiology. Part A, Molecular & integrative physiology 140 (4),

pp. 387–395. DOI: 10.1016/j.cbpb.2005.02.001.

LUNDBERG, J. L.; MOONEY, E. J.; ROGERS, C. E. (1969): Diffusion and solubility of

methane in polyisobutylene. In J. Polym. Sci. A-2 Polym. Phys. 7 (5), pp. 947–962.

DOI: 10.1002/pol.1969.160070517.

MASUDA, M.; SAITO, Y.; IWANAMI, T.; HIRAI, Y. (1992): Effects of hydrostatic pressure

on packaging materials for food. In Claude Balny, Rikimaru Hayashi, Karel

Heremans, Patrick Masson (Eds.): High pressure and biotechnology, vol. 224. 224

volumes. Montrouge, France: John Libbey EUROTEXT (European Seminar on High

Pressure and Biotechnology), pp. 545–547.

MERCEA, P. (2008): Models for diffusion in polymers. In O. G. Piringer, A. L. Baner

(Eds.): Plastic packaging. Interactions with food and pharmaceuticals. 2nd ed.

Weinheim, Chichester: Wiley-VCH, pp. 123–161.

MICHAELS, A. S.; BIXLER, H. J. (1961): Solubility of gases in polyethylene. In J. Polym.

Sci. 50 (154), pp. 393–412. DOI: 10.1002/pol.1961.1205015411.

MUTH, O.; HIRTH, T.; VOGEL, H. (2001): Investigation of sorption and diffusion of

supercritical carbon dioxide into poly(vinyl chloride). In The Journal of Supercritical

Fluids 19 (3), pp. 299–306.

NAITO, Y.; KAMIYA, Y.; TERADA, K.; MIZOGUCHI, K.; WANG, J.‐S. (1996): Pressure

dependence of gas permeability in a rubbery polymer. In Journal of Applied Polymer

Science 61 (6), pp. 945–950.

NAITO, Y.; MIZOGUCHI, K.; TERADA, K.; KAMIYA, Y. (1991): The effect of pressure on

gas permeation through semicrystalline polymers above the glass transition

temperature. In Journal of Polymer Science Part B: Polymer Physics 29 (4),

pp. 457–462.

NILSSON, F.; HALLSTENSSON, K.; JOHANSSON, K.; UMAR, Z.; HEDENQVIST, M. S.

(2013): Predicting solubility and diffusivity of gases in polymers under high


71

pressure: N2 in polycarbonate and poly(ether-ether-ketone). In Ind. Eng. Chem. Res.

52 (26), pp. 8655–8663. DOI: 10.1021/ie300975h.

PAPADOPOULOU, V.; ECKERSLEY, R. J.; BALESTRA, C.; KARAPANTSIOS, T. D.; TANG,

M.-X. (2013): A critical review of physiological bubble formation in hyperbaric

decompression. In Advances in Colloid and Interface Science 191–192 (0), pp. 22–

30. DOI: 10.1016/j.cis.2013.02.002.

PATERSON, M. S. (1964): Effect of pressure on young’s modulus and the glass transition

in rubbers. In J. Appl. Phys. 35 (1), p. 176. DOI: 10.1063/1.1713063.

PINO, M.; DUCKETT, R. A.; WARD, I. M. (2005): Single and mixed gas diffusion

through polyethylene films. In Polymer 46 (13), pp. 4882–4890. DOI:

10.1016/j.polymer.2005.02.118.

PRAUSNITZ, J. M.; LICHTENTHALER, R. N.; AZEVEDO, E. G. d. (1999): Molecular

thermodynamics of fluid-phase equilibria. 3rd ed. Upper Saddle River, N.J.: Prentice

Hall PTR (Prentice-Hall international series in the physical and chemical engineering

sciences).

RAMESH, N. S.; RASMUSSEN, D. H.; CAMPBELL, G. A. (1991): Numerical and

experimental studies of bubble growth during the microcellular foaming process. In

Polym Eng Sci 31 (23), pp. 1657–1664. DOI: 10.1002/pen.760312305.

RAMSING, N.; GUNDERSEN, J. (2000): Seawater and Gases. Tabulated physical

parameters of interest to people working with microsensors in marine systems.

Edited by Unisense. Denmark.

RICHTER, T. (2010): Machine and method for packaging and high pressure treatment of

products. App. no. US 2011/0067355 A1.

RICHTER, T. (2011): Der Einfluss hohen hydrostatischen Drucks auf polymere

thermoplastische Lebensmittelverpackungen. München, Techn. Univ., Diss., 2011.



5/15/2014. Patent no. US8771773B2.

RU-TING, X.; XING-YUAN, H. (2015): Predictive calculation of carbon dioxide solubility

in polymers. In RSC Adv 5 (94), pp. 76979–76986. DOI: 10.1039/C5RA15109K.


72

SANDER, R. (2015): Compilation of Henry’s law constants (version 4.0) for water as

solvent. In Atmos. Chem. Phys. 15 (8), pp. 4399–4981. DOI: 10.5194/acp-15-4399-

2015.

SARRASIN, F.; MEMARI, P.; KLOPFFER, M.-H.; LACHET, V.; TARAVEL CONDAT, C.;

ROUSSEAU, B.; ESPUCHE, E. (2015): Influence of high pressures on CH4, CO2 and

H2S solubility in polyethylene. Experimental and molecular simulation approaches

for pure gas and gas mixtures. Modelling of the sorption isotherms. In Journal of

Membrane Science 490, pp. 380–388. DOI: 10.1016/j.memsci.2015.04.040.

SATO, Y.; FUJIWARA, K.; TAKIKAWA, T.; Sumarno; TAKISHIMA, S.; MASUOKA, H.

(1999): Solubilities and diffusion coefficients of carbon dioxide and nitrogen in

polypropylene, high-density polyethylene, and polystyrene under high pressures and

temperatures. In Fluid Phase Equilibria 162 (1-2), pp. 261–276. DOI:

10.1016/S0378-3812(99)00217-4.

SATO, Y.; TAKIKAWA, T.; SORAKUBO, A.; TAKISHIMA, S.; MASUOKA, H.; IMAIZUMI, M.

(2000): Solubility and diffusion coefficient of carbon dioxide in biodegradable

polymers. In Ind. Eng. Chem. Res. 39 (12), pp. 4813–4819. DOI: 10.1021/ie0001220.

SATO, Y.; TAKIKAWA, T.; TAKISHIMA, S.; MASUOKA, H. (2001): Solubilities and

diffusion coefficients of carbon dioxide in poly(vinyl acetate) and polystyrene. In

The Journal of Supercritical Fluids 19 (2), pp. 187–198. DOI: 10.1016/S0896-

8446(00)00092-9.

SATO, Y.; YURUGI, M.; FUJIWARA, K.; TAKISHIMA, S.; MASUOKA, H. (1996):

Solubilities of carbon dioxide and nitrogen in polystyrene under high temperature

and pressure. In Asian Thermophysical Properties Conference 125 (1–2), pp. 129–

138. DOI: 10.1016/S0378-3812(96)03094-4.

SCHEICHL, R.; KLOPFFER, M.-H.; BENJELLOUN-DABAGHI, Z.; FLACONNÈCHE, B. (2005):

Permeation of gases in polymers: parameter identification and nonlinear regression

analysis. In Journal of Membrane Science 254 (1-2), pp. 275–293.

SCHNITZLER, J. von; EGGERS, R. (1999): Mass transfer in polymers in a supercritical

CO2-atmosphere. In The Journal of Supercritical Fluids 16 (1), pp. 81–92.


73

SCHRITTESSER, B.; PINTER, G.; SCHWARZ, T.; KADAR, Z.; NAGY, T. (2016): Rapid gas

decompression performance of elastomers – A study of influencing testing

parameters. In Procedia Structural Integrity 2, pp. 1746–1754. DOI:

10.1016/j.prostr.2016.06.220.

SHERIDAN, M.H.; SHEA, L.D.; PETERS, M.C.; MOONEY, D.J. (2000): Bioabsorbable

polymer scaffolds for tissue engineering capable of sustained growth factor delivery.

In Journal of Controlled Release 64 (1-3), pp. 91–102. DOI: 10.1016/S0168-

3659(99)00138-8.

SPYCHER, N.; PRUESS, K.; ENNIS-KING, J. (2003): CO2 - H2O mixtures in the

geological sequestration of CO2. I. Assessment and calculation of mutual solubilities

from 12 to 100°C and up to 600 bar. In Geochimica et Cosmochimica Acta 67 (16),

pp. 3015–3031.

SPYRIOUNI, T.; BOULOUGOURIS, G. C.; THEODOROU, D. N. (2009): Prediction of

sorption of CO2 in glassy atactic polystyrene at elevated pressures through a new

computational scheme. In Macromolecules 42 (5), pp. 1759–1769. DOI:

10.1021/ma8015294.

STERR, J.; FLECKENSTEIN, B. S.; LANGOWSKI, H.-C. (2015a): The effect of high

pressure processing on tray packages with modified atmosphere. In Food

Engineering Reviews 7 (2), pp. 209–221. DOI: 10.1007/s12393-014-9081-z.

STERR, J.; RÖTZER, K.; WECK, K.; WIRTH, A. L. K.; FLECKENSTEIN, B. S.; LANGOWSKI,

H.-C. (2015b): In-situ measurement of oxygen concentration under high pressure and

the application to oxygen permeation through polymer films. In J. Chem. Phys 143

(11), p. 114201. DOI: 10.1063/1.4931399.

TANG, M.; DU, T.-B.; CHEN, Y.-P. (2004a): Sorption and diffusion of supercritical

carbon dioxide in polycarbonate. In The Journal of Supercritical Fluids 28 (2-3),

pp. 207–218. DOI: 10.1016/S0896-8446(03)00045-7.

TANG, M.; HUANG, Y.-C.; CHEN, Y.-P. (2004b): Sorption and diffusion of supercritical

carbon dioxide into polysulfone. In J. Appl. Polym. Sci. 94 (2), pp. 474–482. DOI:

10.1002/app.20895.


74

TANG, Q.; YANG, B.; ZHAO, Y.; ZHAO, L. (2007): Sorption and diffusion of

sub/supercritical carbon dioxide in poly (methyl methacrylate). In Journal of

Macromolecular Science, Part B: Physics 46 (2), pp. 275–284.

TROMANS, D. (1998): Temperature and pressure dependent solubility of oxygen in

water: a thermodynamic analysis. In Hydrometallurgy 48 (3), pp. 327–342.

TSUJITA, Y. (2003): Gas sorption and permeation of glassy polymers with microvoids.

In Progress in Polymer Science 28 (9), pp. 1377–1401.

VAN LIEW, H. D.; BURKARD, M. E. (1993): Density of decompression bubbles and

competition for gas among bubbles, tissue, and blood. In Journal of Applied

Physiology 75, p. 2293.

VIETH, W. R. (1991): Diffusion in and through polymers. Principles and applications.

Munich, Vienna, New York, Barcelona: Hanser Publishers; Distributed in the USA

and in Canada by Oxford University Press.

VIETH, W. R.; HOWELL, J. M.; HSIEH, J. H. (1976): Dual sorption theory. In Journal of

Membrane Science 1 (2), pp. 177–220. DOI: 10.1016/S0376-7388(00)82267-X.

VRENTAS, J. S.; DUDA, J. L. (1977): Diffusion in polymer—solvent systems. I.

Reexamination of the free‐volume theory. In Journal of Polymer Science: Polymer

Physics Edition 15 (3), pp. 403–416.

VRENTAS, J. S.; VRENTAS, C. M.; DUDA, J. L. (1993): Comparison of Free-Volume

Theories. In Polym J 25 (1), pp. 99–101. DOI: 10.1295/polymj.25.99.

WANG, W.-C. V.; KRAMER, E. J.; SACHSE, W. H. (1982): Effects of high-pressure CO2

on the glass transition temperature and mechanical properties of polystyrene. In J.

Polym. Sci. Polym. Phys. Ed. 20 (8), pp. 1371–1384. DOI:

10.1002/pol.1982.180200804.

WEDLER, G. (1997): Lehrbuch der physikalischen Chemie. 4., völlig überarb. und erw.

Aufl. Weinheim: Wiley-VCH.

WESSLING, M.; BORNEMAN, Z.; VAN DEN BOOMGAARD, T.; SMOLDERS, C. A. (1994):

Carbon dioxide foaming of glassy polymers. In J. Appl. Polym. Sci. 53 (11),

pp. 1497–1512. DOI: 10.1002/app.1994.070531112.


75

WIEBE, R.; GADDY, V. L. (1939): The solubility in water of Carbon Dioxide at 50, 75

and 100°, at pressures to 700 atmospheres. In J. Am. Chem. Soc. 61 (2), pp. 315–318.

DOI: 10.1021/ja01871a025.

WIEBE, R.; GADDY, V. L. (1940): The solubility of Carbon Dioxide in water at various

temperatures from 12 to 40° and at pressures to 500 atmospheres. Critical

phenomena. In J. Am. Chem. Soc. 62 (4), pp. 815–817. DOI: 10.1021/ja01861a033.

WILT, P. M. (1986): Nucleation rates and bubble stability in water-carbon dioxide

solutions. In Journal of Colloid and Interface Science 112 (2), pp. 530–538.

YAMABE, J.; NISHIMURA, S. (Eds.) (2010): Estimation of critical pressure of

decompression failure of EPDM composites for sealing under high-pressured

hydrogen gas. Proceedings of the 18th European Conference on Fracture.

ZHANG, Z.; HANDA, Y. P. (1998): An in-situ study of plasticization of polymers by

high-pressure gases. In Journal of Polymer Science Part B: Polymer Physics 36 (6),

pp. 977–982.

76

PUBLICATION II CHAPTER 3

77

3 The effect of high pressure processing on the integrity of

polymeric packaging – Analysis and categorization of

occurring defects

The fact that most of the food products are ready packed before high pressure treatment

raises the question whether the packaging still fulfils all necessary requirements after

processing. In this review manuscript investigations concerning the structural behaviour

and the stability of packaging materials under high pressure processing were considered

and it was getting clear that often damages and alterations of the polymeric structure

occur. Common damage mechanisms of high pressure-treated packaging materials are

defined and classified in this review. These effects are allocated to the physicochemical

impact that created them. The effects were separated into direct effects induced by the

action of the high pressure alone and indirect effects that are mediated via compressed

contents of the package, i.e. filled product and gaseous headspace. The direct effects were

split up in reversible and irreversible structural changes. Reversible structural changes

could be only detected indirectly via in-situ measurements of transport processes under

high pressure. Results concerning the permeation, diffusion and solubility of aroma

compounds in different polymers are presented. Irreversible changes, like changes in gas

permeation, alterations in crystallinity or density or macroscopic damages (e.g.

delamination) were listed. The indirect effects are mainly generated by compressed

headspace gases, the concomitant heat of compression and the consequences of increased

amounts of gases dissolved in the polymers. Topics regarding strain induced

crystallisation, local thermal effects, solution effects, extraction, localised decompression

failures and sorption induced plasticization and crystallization are discussed. If

applicable, current theoretical approaches have been allocated to the different categories

of damage.


78

The effect of high pressure processing on the integrity of

polymeric packaging – Analysis and categorization of

occurring effects

BENEDIKT STEFAN FLECKENSTEIN1, JULIA STERR1, HORST-CHRISTIAN LANGOWSKI1,2



Published online 21 February 2013 in Wiley Online Library1: Packaging Technology

and Science (2014); Volume 27, Issue 2, pp. 83–103; DOI: 10.1002/pts.2018

ABSTRACT

As high pressure processing is used increasingly for the treatment of packed products,

different packaging has been investigated with respect to their structural behaviour and

stability under high pressure processing. Often, failures and changes of the polymeric

structure occur. Common damage symptoms of high pressure-treated packaging materials

are defined and classified in this review. These damage symptoms are allocated to the

physicochemical effects that created them. The effects may be separated into direct effects

induced by the action of the high pressure alone and indirect effects that are mediated via

compressed contents of the package, i.e. filled product and gaseous headspace. The direct

effects split up again in reversible and irreversible structural changes. The indirect effects

are generated by compressed headspace gases, other compressed substances and the

consequences of increased amounts of gases dissolved in the polymers. If applicable,

current theoretical approaches have been allocated to the different categories of damage.

1 Permission for the re-use of the article is granted by John Wiley & Sons, Ltd.


79

Content Publication II

3 The effect of high pressure processing on the integrity of polymeric packaging –

Analysis and categorization of occurring defects ............................................................ 77

3.1 Introduction ...................................................................................................... 80

3.2 Direct effects .................................................................................................... 81

3.2.1 Microscopic structural changes in the nanometre to micrometre range ... 81

3.2.1.1 Reversible structural changes ............................................................ 81

3.2.1.2 Models to describe the high pressure effect on sorption, diffusion and

permeation 85

3.2.1.3 Irreversible structural changes: changes in crystallinity ................... 86

3.2.2 Changes of layers and layer structure ....................................................... 92

3.3 Indirect effects ................................................................................................. 94

3.3.1 Gas compression ....................................................................................... 94

3.3.2 Influence of non-gaseous substances ...................................................... 104

3.4 Conclusion ..................................................................................................... 105

3.5 Publication bibliography ................................................................................ 107


80

3.1 Introduction

“The real value of packaging is that the package is an integral part of the product today”

(AHMED ET AL. 2005). This is naturally in force for packaged and high pressure processed

food. The visual and structural integrity of a packaging material after high pressure

processing (HPP) is a very important criterion for its selection. First, the consumer

demands an intact, visually undamaged appearance. Second, the structural integrity of

packaging material is the basis of its protective effect. This is not easy to achieve in the

case of flexible vacuum packages. For instance, high pressure treatment may affect gas

or water vapour permeability and thus hamper packaging integrity (JULIANO ET AL. 2010).

Instead of flexible vacuum packages, the packaging industry more often asks for tray

packages, comprising a gaseous headspace filled with modified atmosphere for reasons

of better handling, of a more attractive appearance and of the additional protective effect

of gases such as CO2.

Generally, the developments in different disciplines of HPP research for food

preservation have been extensively reviewed (BALASUBRAMANIAM, FARKAS 2008;

BARBOSA-CÁNOVAS, JULIANO 2008; BERMÚDEZ-AGUIRRE, BARBOSA-CÁNOVAS 2011;

CANER ET AL. 2004a; GALIĆ ET AL. 2011; GUILLARD ET AL. 2010; GUPTA,

BALASUBRAMANIAM 2011; JUNG ET AL. 2011; MIN, ZHANG 2007; MORRIS ET AL. 2007a;

NORTON, SUN 2008; RASTOGI ET AL. 2007; SAN MARTÍN ET AL. 2002; TOEPFL ET AL. 2006;

TORRES, VELAZQUEZ 2005). For plastic packaging for food in combination with HPP, an

extensive review has recently been published by JULIANO ET AL. (2010). The paper mainly

deals with the general requirements for pouches in HPP applications. The authors focus

on industrial standards, q.v. (LAMBERT ET AL. 2000), on the Commission directive

90/128/EEC (The European Commission 2/23/1990) and on the overall packaging

integrity on the basis of barrier and mechanical characteristics. High pressure-induced

damages of the packaging materials were not treated in detail. So, the aim of the present

paper is to give an overview of the different damages as induced by HPP and the related

mechanisms in view of the different packaging materials.

In the following, the current state of knowledge of the effects of HPP on packaging

materials is presented. The observed damages have been systematized with regard to their

origin and the mechanisms of their generation (Figure 3-1). The main principle behind

this scheme is to separate between direct high pressure-induced effects and indirect


81

effects, which are mainly created by the compressed contents of the packages, i.e.

products and gaseous headspace.

3.2 Direct effects

In this paper, we define direct effects as the immediate consequences of the HPP on the

polymer. As the schematic representation shows (Figure 3-1), we differentiate these in

microscopic structural changes in the nanometre to micrometre range and changes of

layers and layer structures.

3.2.1 Microscopic structural changes in the nanometre to micrometre range

One group of immediate consequences of the HPP on polymers is represented by

structural changes in the nanometre to micrometre range. The detectable effects of these

consequences may be reversible or irreversible. The effects themselves may be detected

directly by structural investigations or indirectly by studies of the permeation/diffusion

properties of different substances in the polymers during HPP.

3.2.1.1 Reversible structural changes

Reversible structural changes cover alterations where the polymeric structure changes

during the high pressure treatment and returns to its original state after the process. Such

behaviour asks for measurement methods that may be carried out in situ, i.e. during the

HPP. So far, for hydrostatic pressures >50MPa, this kind of effects has only been

observed with the help of the permeation of flavour substances. The following studies

show that the permeation and the absorption of specific liquid substances are significantly

reduced even at moderate hydrostatic pressures. An overview on observed reversible

structural changes is given in Table 3-1.

Recently, RICHTER ET AL. (2010) could show directly by in situ studies that during high

pressure treatment, the permeation coefficients of benzoic acid through polyamide 6

(PA6) and of β-ionone through low-density polyethylene (PE-LD) decrease reversibly

upon an increase of pressure during the treatment. After pressure release, the permeability

returns to its initial value within the accuracy of the measurement. Similar findings were

published by SCHMERDER ET AL. (2005) for the solution and diffusion of raspberry ketone

through PA6.


82

Figure 3-1 Schematic representation of the effects of high pressure on polymeric packaging materials

Other papers report on more indirect experiments like GÖTZ, WEISSER (2002) who

compared the in situ measured concentration of p-cymene permeating into a solution

outside a bag over time at 0.1 and 50 MPa. In this case, the amount of p-cymene outside

the bag increased slower at 50 MPa than at 0.1 MPa, indicating a reduction of the

permeation coefficient. KÜBEL ET AL. (1996) could show that “high hydrostatic pressure

raised the diffusion barrier in the polymers” for PE-LD/high-density polyethylene (PE-

HD)/PE-LD, PE-HD and polyethylene terephthalate (PET)/aluminium (Al)/PE-LD. The

experimental set-up in their study was an inner bag, filled with p-cymene, which was

packed in an outer bag, filled with an ethanol/water solution. Their findings on PE-

LD/PE-HD/PE-LD films show that a lower amount of p-cymene permeates out of the

inner bag in the case of pressurized samples and that the latter absorb only one third of p-

cymene at 500 MPa in comparison with untreated samples. These findings are in concert

with the results of LUDWIG ET AL. (1994). They measured the change of p-cymene

concentrations before and after HPP also via a bag-in-bag method.


83

Table 3-1 Reversible structural effects: decrease of diffusion or permeation coefficients of flavour substances

Polymer Treatment Thickness Permeate Method Effect of high pressure Reference

PA6 0.1-200 MPa; 23 or 60°C

20 µm Benzoic acid; Carvacrol; Rasperry ketone

UV/Vis in situ

P * is reduced by a factor of up to 7 (Benzoic acid, 23°C)

(RICHTER ET

AL. 2010) PE-LD 50 µm β-ionone P is reduced by a factor of ~ 125

PA6 0.1-400 MPa; 23 or 60°C

55 µm Rasperry ketone in situ P is reduced by a factor of ~ 4 (40°C) (SCHMERDER

ET AL. 2005)

PE-LD/PE-HD/PE-LD 0.1 and 50 MPa; 23°C

12/12/12 µm p-Cymene UV/Vis; in situ bag in bag

Reduced permeation P50MPa ~ 2.7 *10-13 mg*m/s*m2*Pa P0.1MPa= n/a

(GÖTZ, WEISSER

2002)

PET/Al/PE-LD

0.1-450 MPa; 25°C; 60 min

8/7/75 µm

p-Cymene; Acetophenone

UV/Vis; bag in bag

Less decrease in the inner bag (p-Cymene) No diffusion of

Acetophenone neither through nor into the film at 450 MPa

(KÜBEL ET AL. 1996)

PE-LD/PE-HD/PE-LD 12/12/12 µm ~ 2/3 less sorption (p-Cymene)

PE-HD 14 µm ~ 40% less sorption (p-Cymene)

PE-LD 500 MPa; 25°C; 24h

100 µm

p-Cymene UV/Vis; bag in bag

~50% less decrease in the inner bag (LUDWIG ET

AL. 1994) PE-LD/PE-HD/PE-LD 12/12/12 µm ~35% less decrease in the inner bag

PET/Al/PE-LD 500 MPa; 25°C; 60h

8/7/75 µm Slight decrease of loss by sorption, no permeation

PE-LD 800 MPa; 20-40°C; 5 min

50 µm 2-Hexanone; 2-Heptanol; Ethyl hexanoate; D-Limonene; n-pentane; Dichlormethane; Ethyl butanoate; Acetic acid

Filled with food simulating liquid; Extraction from film and solution Detection via gas chromatography

Slight increase of the total amount (~15%; out of water, 3% acetic acid); no change (out of 15% ethanol)

(MAURICIO-IGLESIAS ET

AL. 2011)

PLA 40 µm Decrease of the total amount (~ 65%)

PE-LD 800 MPa; 90°C-115°C; 5 min

50 µm Strong rise of total amount of sorbed aroma compounds (up to 350% out of water)

PLA 40 µm Increase of total amount of sorbed aroma compounds (up to 120% out of 15%ethanol)

UV/Vis, Ultraviolet-visible spectroscopy; n/a, not applicable. * P = Permeation coefficient


84

Whereas the aforementioned authors measured the amount of substances permeating

through polymers after a time of at least 25 min, MAURICIO-IGLESIAS ET AL. (2011)

measured the uptake of several flavouring substances (Table 3-1) by polylactide (PLA)

after a treatment of 7 min at 0.1 MPa and 5 min at 800 MPa to reproduce a realistic

treatment time. They detected a reduced uptake of flavour out of a food-simulating liquid

by PLA, if treated at temperatures below Tg. Furthermore, they showed that the uptake of

flavour by PE-LD during the time of high pressure treatment depends on the food-

simulating liquid.

MAURICIO-IGLESIAS ET AL. (2011) also tested the consequences of combined high

pressure/high temperature on the solubility of substances in polymers. In this case,

temperatures above Tg (90 to 115°C) increased the uptake of flavour substances by PE-

LD and PLA.

For completeness, it is to be mentioned that in a study by CANER ET AL. (2004b), pouches

made from single layer polypropylene (PP) (25 µm) and PE/PA6/ethylene-vinyl alcohol

copolymer (EVOH)/PE (62.5 µm) did not show significant changes in the sorption

behaviour during high pressure treatment. A study by CAMPION, MORGAN (1992) deals

with the high pressure permeation of gases in polymers. Even if the test set-up implied a

pressure gradient (42MPa to ‘low pressure’) and the decompression was not recorded, the

occurring decrease of the permeation coefficient indicates that reversible effects may also

occur at the permeation of gases.

Whereas these in situ studies of gas permeation at high pressures still need further

attention, there are many papers that deal with the gas or vapour permeability through

polymers after high pressure treatment. Most of them have been reviewed by JULIANO ET

AL. (2010). Recent publications about this topic that were not reviewed in the paper of

Juliano et al. are for example: (BULL ET AL. 2010; DHAWAN ET AL. 2011; GALOTTO ET AL.

2009; GALOTTO ET AL. 2010; LARGETEAU 2010; SANSONE ET AL. 2012).


85

3.2.1.2 Models to describe the high pressure effect on sorption, diffusion and

permeation

The free-volume model

The observation of a temporary change of the permeation coefficient can be explained

via the free-volume model for diffusion in polymers (RICHTER ET AL. 2010). The model

is based on the assumption that within or throughout an ordered matrix, there are discrete

cavities, which may be either fixed or mobile. Dissolved molecules may only move when

there is a free space to jump in. The polymer chains, which form the cavities, are subject

to the Brownian motion. In case that one of the formed cavities is large enough to contain

the diffusing molecule and the molecule possesses enough energy to jump in, movements

occur, cf. (BARRER 1951; VIETH 1991; VRENTAS, DUDA 1977; ZIELINSKI, DUDA 1992). A

bibliographic review about common theories of transport properties of gases in polymers

at moderate pressure was written inter alia by KLOPFFER, FLACONNÈCHE (2001). They

show that the permeability of gases and organic vapours depends on the polymer

structure, on the penetrant size and on the conditions of pressure and temperature.

However, SCHEICHL ET AL. (2005) pointed out that for gas–polymer systems “an increase

of the hydrostatic pressure leads to a competition between two phenomena with opposite

effect”:

• The free volume inside the polymer matrix may be reduced because the polymer

density increases via polymer compression.

• The free volume inside the polymer matrix may be increased because the

macromolecular chains in the polymer are plasticized by an increase of the gas

concentration.

Even though many substances may act as plasticizers (see the Section ’Sorption-induced

plasticization and crystallization’), according to the findings of the aforementioned

authors, the first of these phenomena appears to predominate at least at p-cymene and

acetophenone in combination with PE.

Gas and vapour diffusion at pressures over 100 MPa has not yet been studied in the

literature. Studies at CO2 pressures up to 100 MPa indicate that the permeability of a

polymer depends on the gas–polymer interaction regarding plasticization, diffusivity,

sorption and others: (CAMPION, MORGAN 1992; HOUDE ET AL. 1992; JORDAN ET AL. 1989;


86

SADA ET AL. 1988; SANDERS 1988). For a better understanding of diffusion processes at

high pressure, it would be essential to measure the permeation of both gases and vapours

at high pressure in situ.

‘Dual sorption’ model

It is well known that sorption is the first step of permeation. If Henry’s law is no longer

applicable, another approach to describe the solubility is the dual sorption model. In this

model, Henry’s-type sorption and Langmuir-type sorption occur simultaneously. As a

function of the external pressure of the penetrant, the resulting overall concentration of

the penetrator can be calculated (ANGELIS, SARTI 2011; VIETH 1991). In the literature,

however, most papers deal with pressures below 70 MPa. In this pressure range, the

solubility of molecules in polymers increases with rising partial pressure, e.g. (BOYER,

GROLIER 2005b; CHANDRA, KOROS 2009; ESLAMI, MÜLLER-PLATHE 2007; PATERSON ET

AL. 1999; SOLMS ET AL. 2005). It has to be said that at higher pressures, common

headspace gases are in their supercritical state. This generally will lead to a higher

solubility (cf. overview of critical temperature and pressure in Table 3-4 and Chapter

‘Solution effects’).

For liquid penetrants, only two groups report on solubility measurements at higher

pressures, showing that the solubility decreases at a pressure range above several hundred

MPa (KÜBEL ET AL. 1996; LUDWIG ET AL. 1994).

3.2.1.3 Irreversible structural changes: changes in crystallinity

The question whether high pressure treatment directly affects the crystallinity of polymers

cannot be answered unambiguously. In the literature, different trends are to be found in

the field of high pressure-induced irreversible changes in the crystallinity of polymers,

depending on the nature of the polymer. There is still an inconsistency between different

methods to determine the crystallinity, e.g. Fourier transform infrared spectroscopy (FT-

IR), Raman spectroscopy, differential scanning calorimetry (DSC), X-ray diffraction

(XRD) and small angle X-ray scattering, e.g. (GRUVER ET AL. 2000; RUNT,

KANCHANASOPA 2004). LÓPEZ-RUBIO ET AL. (2003) mentioned the “complex issue of

rigorous crystallinity determination for [...] polymers”. Because of these facts, it is

essential that additional reference tests should always be carried out with pure monolayer

materials to gain a common background.


87

In the following, the related papers are dedicated to three subcategories: in findings that

indicate differences due to the measurement method, in papers that report on multilayer

with EVOH and in unspecific studies. An overview of reported effects is shown in Table

3-2.


88

Table 3-2 Irreversible structural changes: changes in crystallinity.

Polymer Treatment Method Changes in crystallinity References

PE-LD 200 to 800 MPa 5/10 min 25/75°C

XRD crystallinity increased with increasing pressure from ~42% (control) to ~58% (800 MPa-5 min-75°C Time no effect

(YOO ET AL. 2009)

DSC No significant changes

PLA 650 MPa 15 min 22 to 26°C

DSC Decrease from ~20% to ~12% (L-isomer) (AHMED ET AL. 2009) FT-IR No change observed

PET/EVOH32/PP 680 MPa 3/5 min 105/100°C

DSC XRD

No change of Tm and ∆H (DSC) Decrease in the overall crystallinity (XRD) (DHAWAN ET AL.

2011) PET/PP/tie/ PA6/EVOH27/PA6/tie/PP

No change of Tm and ∆H (DSC) Slight increase of crystallinity (XRD)

PP/EVOH26/PP 400/800 MPa 5/10 min 40/75°C

FT-IR Increase of crystallinity

(LÓPEZ-RUBIO ET

AL. 2005)

DSC WAXS, SAXS Certain improvement

PP/EVOH48/PP FT-IR

No change of crystallinity DSC WAXS, SAXS Small changes

PA6/EVOH/PE 690 MPa; 10 min; 95°C

DSC No significant differences (SCHAUWECKER

ET AL. 2002) PA6/EVA

600 MPa 10 min 110°C

DSC

EVOH: significant decrease of ∆H (AYVAZ ET AL. 2012)

PA6/EVOH/EVA PA6, EVA: slight, but not statistically decrease (p>0.05) of ∆H

PETmet/PE PE: significant decrease of ∆H PET: no change of ∆H

PET/AlOx 500 MPa 15 min 50°C

DSC

Increase from 28,5% to 36.3% (oil simulant) and 35.5% (aqueous simulant) (GALOTTO ET AL.

2009) PLA/SiOx-PLA

Decrease from 2.4% to 1.9% + redistribution (aqueous simulant) Decrease from 2.4% to 1.7% (oil simulant)

PLA/PLA/PLA (multilayer) 700 MPa; 15 min; 40°C

DSC

Untreated crystallinity: 25.3-25.8% (in each case multilayer average) Crystallinity (tap water): 24.2-25.4% Crystallinity (solid carrots): 26.4-26.9% Crystallinity (carrot juice): 25.2-25.7%

(SANSONE ET AL. 2012)


89

Polymer Treatment Method Changes in crystallinity References

700 MPa; 15 min; 90°C

Untreated crystallinity: 25.3-25.8% (in each case multilayer average) Crystallinity (tap water): 29.2-29.6% Crystallinity (solid carrots): 29.3-29.7%

PA6 50 to 600 MPa 23°C

DSC No change (SCHMERDER ET

AL. 2005)

PET/PE-LD 600 MPa 30°C

Raman Increase (not specified) Failure analysis - white spot

(RICHTER ET AL. 2010)

PP/PET

695 MPa 10 min 70 to 86°C release 2s

SAXS Increase Strain-induced (see Section on ‘Indirect effects: strain induced crystallization’)

(FAIRCLOUGH, CONTI 2009)

Tested layer; WAXS, wide-angle X-ray scattering; SAXS, small-angle X-ray scattering. ∆H= heat of fusion; numbers following EVOH give ethylene content in mol%.


90

Differences due to measurement method (changes in crystallinity)

As previously mentioned, discrepancies in the changes in crystallinity obtained by

different measurement methods were reported by some authors. YOO ET AL. (2009)

described the influence of DSC and XRD as methods to characterize crystallinity on the

findings. In their study on PE-LD film material, they investigated the effect of different

HPP conditions (200–800 MPa; 5/10 min; 25/75°C) on the degree of crystallinity. The

results of the DSC analysis indicate that HPP does not affect the crystallinity of the

specimens significantly, whereas the XRD results suggest an increase of crystallinity

from 42% to 58%. In addition, the XRD data show that the crystallinity increased with

increasing pressure from 200 to 600 MPa, whereas the processing time had no significant

consequence on the crystallinity. Similar discrepancies between the findings obtained by

DSC and XRD were found by DHAWAN ET AL. (2011) (see Section on ‘Influence of

ethylene content’). A similar difference between the results obtained by two methods was

detected for PLA by AHMED ET AL. (2005): In this case, DSC indicates a decrease of the

degree of crystallinity from 20% to 12% at a pressure rise of 650 MPa (for 15 min at 22–

26°C), whereas the crystalline bands of the FT-IR show no changes.

YOO ET AL. (2009) interpreted the differences of the methods as a result of the

dependencies on different physical effects. In XRD measurements, crystallinity is

obtained from the volume fraction of different structures, whereas DSC measures

crystallinity as a function of the heat of fusion. So, when the volume of a polymer is

reduced by HPP, cf. (BERET, PRAUSNITZ 1974; KOVARSKII 1994; SCHAUWECKER ET AL.

2002), XRD will produce higher values. In the application of DSC, however, the heat of

fusion obtained has to be compared with the heat of fusion of a perfect crystal (100%

crystallinity). At this, it is difficult to construct the correct ‘peak baseline’ for the

integration (SCHICK 2009).

Influence of ethylene content of EVOH on changes in crystallinity.

In some publications, the authors report on an influence of the ethylene content of EVOH

on high pressure-induced changes of crystallinity. In the case of EVOH with 26%

ethylene content in a multilayer structure (PP/EVOH26/PP), a slight increase in

crystallinity could be detected, whereas for EVOH with an ethylene content of 48% in a

multilayer structure (PP/EVOH48/PP), no increase in crystallinity was observed at equal

process conditions (400/800 MPa; 5/10 min; 40/75°C). This was measured by FT-IR,


91

DSC, wide angle X-ray scattering and small angle X-ray scattering (LÓPEZ-RUBIO ET AL.

2005).

In a recent paper, DHAWAN ET AL. (2011) have tested two films (PET/EVOH32/PP and

PET/PP/tie/PA6/EVOH27/ PA6/tie/PP) with DSC and XRD after pressure-assisted

thermal processing (600 MPa; 3/5 min; 105/100°C). The DSC results show no influence

of the pressure-assisted thermal processing on the thermal characteristics (Tm and ΔH) of

the EVOH layer. In contrast, the XRD measurements show a decrease in the overall

crystallinity of the former film and a slight increase of crystallinity of the latter film. The

increase in crystallinity at EVOH27 shows similarities with the findings of LÓPEZ-RUBIO

ET AL. (2005).

Unfortunately, also papers without specifications for the content of ethylene in EVOH

are to be found: SCHAUWECKER ET AL. (2002) analysed a multilayer film with an EVOH

core (PA6/EVOH/PE). In their case, no changes in crystallinity fraction could be detected

by DSC measurements (690 MPa; 10 min; 95°C). AYVAZ ET AL. analysed a

PA6/EVOH/ethylene-vinyl acetate copolymer (EVA) multilayer after pressure-assisted

thermal processing (600MPa; 10min; 110°C). Here, the DSC shows a significant decrease

of the heat of fusion (ΔH) of EVOH.

Unspecific studies

A small decrease in crystallinity from 2.4% to 1.7% (oil simulant) respectively to 1.9%

(aqueous simulant) was found by GALOTTO ET AL. (2009) when they analysed high

pressure-treated PLA with DSC. But the analysed film (PLA/SiOx/PLA) had a very low

initial crystallinity of 2.4% before HPP (500 MPa; 15 min; 50°C), and the changes are

within the error of measurement. Furthermore, the results of the DSC point out that there

was a redistribution in the crystallites size when PLA is in contact with a liquid of a great

affinity to PLA (in this case an aqueous simulant). In this study, GALOTTO ET AL. also

give an account of a large increase in the degree of crystallinity (from 28.5% to ~36%;

DSC) in PET after HPP treatment (500 MPa; 15 min; 50°C).

By contrast, in a study on the suitability of PLA films for HPP, SANSONE ET AL. (2012)

detected an increase of crystallinity by DSC (700 MPa; 15 min; 40–110°C). In this case,

the magnitude of the increase depended on temperature and the packed food and was

about 4% for a treatment at 110°C. No variation of crystallinity through HPP (50–600


92

MPa; 23°C) was found by DSC in PA6 (SCHMERDER ET AL. 2005). AYVAZ ET AL. (2012)

analysed EVOH in combination with PA6, EVA, PET and PE thermally. The heats of

fusion of PA6, EVA and PET show no significant decrease after high pressure treatment

(600 MPa; 10 min; 110°C), whereas the heat of fusion of PE after the same process is

significantly decreased.

In a failure analysis, RICHTER ET AL. (2010) detected an increase of crystallinity in a

damaged area of a PE-LD layer of a PET/PE-LD multilayer at 600 MPa and 30°C by

Raman spectroscopy.

3.2.2 Changes of layers and layer structure

Packaging films with an inorganic layer showed to be affected by HPP in two ways. On

the one hand, the insufficient flexibility of the inorganic material in conjunction with a

high pressure treatment can lead to cracks in the layer. On the other hand, delamination

between a layer and an adjacent polymeric layer can occur because the materials respond

differently to compression and temperature.

Defects generated in inorganic layers

GALOTTO ET AL. reported on defects in inorganic layers caused by HPP (GALOTTO ET AL.

2008; GALOTTO ET AL. 2009; GALOTTO ET AL. 2010). They investigated coated and

laminated films with different coatings on the substrates, namely PET/AlOx,

PLA/SiOx/PLA, PP/SiOx and PETmetallized (met)/PE.

PET/AlOx showed many pinholes and cracks after HPP (500 MPa; 15min; 50°C). It is

noticeable that the damage caused by HPP treatment was less severe when using water as

a food simulant compared with oil as a food simulant. In this case, the damages in the

PET/AlOx film affected all physical properties of the film (modulus of elasticity, tensile

strength and percent elongation). On the contrary, the PLA/SiOx/PLA film showed some

pinholes after HPP treatment independent of the used food simulant. Additionally, bright

areas appear in the film that was in contact with water (GALOTTO ET AL. 2009).

Scanning electron micrographs of PP/SiOx showed that the complete SiOx layer was

destroyed after HPP treatment at 400 MPa and 60°C for 30min (GALOTTO ET AL. 2008).

After the same treatment of PETmet/PE, only cracks could be detected in the layer by

scanning electron microscopy (SEM) (GALOTTO ET AL. 2010). Table 3-3 shows a

summary of all process parameters and occurring damages.


93

GALOTTO ET AL. (2008) and BULL ET AL. (2010) explained the destruction of the inorganic

layer via the insufficient flexibility of the layer material. By nature, cracks will appear if

tensile stresses are greater than the cohesive strength of the coating. Internal stresses will

be generated in most multilayer systems because the materials show different thermo-

mechanical properties (LETERRIER 2003). For example, first cracks appeared in a coating

of 30 nm SiOx on 12 µm PET already at a nominal strain of around 2.5% in the uniaxial

mode (LETERRIER ET AL. 1997).

Delamination of multilayers with inorganic layers or polymeric coatings

Another group of direct effects is the occurrence of delamination of different laminates.

So, a polyester/PA6/aluminium/PP (142 µm) multilayer film shows delamination

between PP/Al when pressurized at ≥200 MPa and ≥90°C for 10 min (SCHAUWECKER ET

AL. 2002). Analogue to these findings, GALOTTO ET AL. (2008) detected delamination in

metallized multilayer polymer. They studied PETmet/PE (75/19 µm) films by visual

inspection and SEM. The parameters of the HPP were 400 MPa for 30 min and at 20±2°C

or 60±2°C.

These findings confirm the results from the study of CANER ET AL. (2003): On scanning

electron micrographs of high pressure-processed (600/800 MPa; 45°C; 5–20 min)

PETmet/ EVA/linear PE-LD, they detected numerous folds and wrinkles on the food

contact side of the film and a delamination between the adjacent PETmet and PE layers

via C-mode scanning acoustic microscopy. In contrast, the multilayer systems of

PET/AlOx/PE-LD and PET/SiOx/PE-LD (and PET/polyvinylidene chloride/EVA/linear

PE-LD) also showed some surface deformation (by SEM) but no indications of a

delamination after an HPP treatment at 800 MPa with 45°C for 20 min. Both

SCHAUWECKER ET AL. (2002) and GALOTTO ET AL. (2008) saw the reason for the

delamination in the different response of metal, polymer and adhesive layers to

compression and temperature.


94

Table 3-3 Damages in inorganic layers.

Polymer Treatment Thickness Description Method Remark Reference

PP/SiOx 400 MPa 20 or 60°C 30 min

20 μm significantly damaged

n/a s. Galotto et al. 2008

(GALOTTO

ET AL. 2010)

PET/AlOx 500 MPa 50°C 15 min

11.6±0.5μm pinholes, cracks SEM

Depends on the food simulating liquid

(GALOTTO

ET AL. 2009)

PLA/SiOx/PLA 35.8/21.5μm pinholes

PETmet/PE 400 MPa 20 or 60°C 30 min

75/19 μm disruption of the coating, delamination wrinkles

vis., SEM

Not affecting mechanical properties

(GALOTTO

ET AL. 2008; GALOTTO

ET AL. 2010) PP/SiOx 21±1 μm

3.3 Indirect effects

Indirect effects arise from the compression of substances and cover a wide field of

reported damages in HPP. The compression of headspace gases, which goes along with a

changeover into the supercritical state of the gas, leads to numerous defects. So, a

localized increase of temperature can cause an alteration of the packaging. A more

complex topic is the influence of solution effects. First, the solubility of other substances

in gas increases. This may induce an extraction of value-adding components out of the

food on the one hand; on the other hand, the supercritical gas can extract constituents out

of the polymers. Second, the concentration of the gases in polymers increases. This in

turn has two effects. Initially, the absorbed gas acts as a plasticizer and can lead to changes

in crystallinity. Furthermore, a too rapid decompression conducts to explosive

decompression failures (blisters) because the specific volume increases too fast. Another

source of damage is the enhanced solubility of liquid substances that can lead to

delamination.

3.3.1 Gas compression

The category ‘gas compression’ contains the consequences of headspace gas compression

caused by a high pressure treatment. It can be divided into three subcategories: Strain-

induced crystallization, local thermal damages caused by the heat of compression and

solution effects caused by the supercritical state of headspace gases during HPP. This


95

implies both the increase of solved gas in a polymer and the increase of solubility of

substances in the supercritical gas.

Strain-induced crystallization

In addition to the named irreversible structural change in the category direct effects,

another kind of these irreversible structural changes is strain-induced crystallization. This

phenomenon has been observed in the form of white lines after HPP (695 MPa; 10 min;

70–86°C) of PP/PET multilayer films (FAIRCLOUGH, CONTI 2009). These lines were

allocated by the authors to both strain induced crystallization and decompression. The

same phenomenon can be found in the papers of GORLIER ET AL. (2001) and FERNANDEZ,

SWALLOWE (2000) and in publications about applications of rapid decompression

(HANDA ET AL. 1999; ZENG ET AL. 2003). The latter will be treated more precisely in the

succeeding section.

Local thermal effects

During high pressure treatment, the phenomenon of adiabatic heating occurs. Because the

product, the headspace gases and also the pressure-transmitting fluids differ in the

emerging heat of compression and the thermal conduction between sample, media and

pressure vessel is different, the temperature distribution in the vessel can be

inhomogeneous (DENYS ET AL. 2000; HARTMANN ET AL. 2003; TING,

BALASUBRAMANIAM 2002). Whereas the heat of compression of pressure-transmitting

fluids, polymeric packaging and liquid foods has been well studied, e.g.

(BALASUBRAMANIAN, BALASUBRAMANIAM 2003; BARBOSA-CÁNOVAS, RODRIGUEZ

2005; BUZRUL ET AL. 2008; KNOERZER ET AL. 2010; PATAZCA ET AL. 2007;

RASANAYAGAM ET AL. 2003; TING, BALASUBRAMANIAM 2002) there are no systematic

investigations on the influence of compressed headspace gas on the packaging. As the

compressive heating effect of gases is much higher than that of liquids, as TING,

BALASUBRAMANIAM (2002) mentioned, a localized increase of temperature by

compressed headspace gases, especially in the corners and angles of a packaging, will

largely influence the damage symptoms . In this regard, RICHTER ET AL. (2010) related

the observed increase in crystallinity in area of white spots to local thermal effects.


96

Solution effects

The solution effects base mainly upon the fact that the headspace gases are in their

supercritical state during high pressure treatment (cf. overview of critical temperatures

and pressures in Table 3-4). This leads to two different impacts. First, the solubility of

substances in the supercritical fluid is enhanced. Second, larger quantities of supercritical

fluid can be solved in polymers. For the HPP of packed food, two potential causes of

defect arise from there: Both negatively impacts on the quality of food by extraction and

localized decompression failures by the oversaturation of the polymers with supercritical

headspace gas.

Table 3-4 Critical temperature and critical pressure of typical headspace gases (EMSLEY 1989; LAX ET AL. 1998).

Gas Critical temperature (K)

Critical pressure (MPa)

Density [STP] (mol/dm³)

Supercritical density (by way of example)

Reference

CO2 304.19 7.38 0.0450

33.9 mol/dm³ at 330 K and 800 MPa 24.2 mol/dm³ at 310 K and 75 MPa

(SPAN, WAGNER

1996)

N2 126.25 3.39 0.0445 10.9 mol/dm³ at 300 K and 31.3 MPa

(DILLER

1983)

O2 154.55 5.05 0.0447 19.6 mol/dm³ at 310 K and 65.3 MPa

(RODERT

1982)

Extraction

The enhanced solubility of certain substances in supercritical fluids is used for extraction

in numerous industrial applications (BRUNNER 2010). This enhanced solubility may have

an impact on two different negative effects in the area of HPP: First, as pressurized CO2

is used, e.g. in supercritical fluid extraction of food components, high pressure treatment

will probably affect the quality of food in some cases. Influencing factors in this

connection are the nature of the extracting solvent, the pressure level, the dwell time and

the solvent: material ratio (JUNG ET AL. 2011). Whereas HPP-induced changes of the

aroma of vacuum-packed food were published recently (RIVAS-CAÑEDO ET AL. 2009;

RIVAS-CAÑEDO ET AL. 2011), negative effects caused by extraction by supercritical

headspace gases are not reported in the literature.

Second, supercritical CO2 can extract additives out of polymers (cf. polymer analysis,

(COTTON ET AL. 1991; VENEMA ET AL. 1993). For example, COTTON ET AL. reported on


97

the extraction of phenol, 2-(5-chloro-2H-benzotriazole-2-yl)6-(1,1-di-methylethyl)4-

methyl (Tinuvin 326) out of PP with supercritical CO2. This effect may thus lead to a

higher migration out of packaging materials. So far, to our knowledge, tests of global and

specific migration of packaging components into food simulants during HPP were only

performed without headspace gases (CANER, HARTE 2005; DOBIÁS ET AL. 2004; LAMBERT

ET AL. 2000; MAURICIO-IGLESIAS ET AL. 2010; MERTENS 1993; OCHIAI, NAKAGAWA

1992). Even though the additional extraction effect may be small and its contribution to

the migration of packaging components low, it should be investigated with regard to the

plastics implementation measure (PIM 10/2011) (Commission Regulation (EU) No

10/2011).

Localized decompression failure

The damages that can be traced back to increased gas solution in polymers are usually

due to a rapid decompression of the solved fluid. The increased solubility of supercritical

fluids in polymers can be seen exemplary in a comparison of the solubility of CO2 in

polysulfone (PSU) at 313 K in Figure 3-2. The chart shows the solubility of CO2 (in

weight%) versus pressure on a double logarithmic scale. The red-dotted vertical line

represents the critical pressure of CO2 (Table 3-4). At an increase of pressure to values

above the critical value, the solubility is increased by a factor of ~100.

TANG ET AL. (2007) looked at the sorption isotherms of CO2 in the transition region of

gaseous to supercritical state (up to 15 MPa) in poly(methyl methacrylate) (PMMA) at

different temperatures. Thereby, every sorption isotherm shows a turning point for the

sorption saturation, i.e. below the turning point the sorption saturation increases slightly

with pressure, whereas the increase of the sorption saturation is steeper with pressures

above the turning point. Interestingly, the turning point can also be below the critical

pressure of CO2, depending on the temperature.

To our knowledge, there is no experimental data of the solubility of gases in polymers at

very high pressure (PATERSON ET AL. 1999). FAIRCLOUGH, CONTI (2009) conducted a

rough calculation of the solubility of N2 in PP at pressures up to 700 MPa. According to

their calculation, the solubility of nitrogen in PP increases by a factor of 36.3 upon a

pressure rise from 0.1 to 700 MPa.


98

At the decompression during high pressure treatment, the solved supercritical gas will

expand to avoid an oversaturation and balance the thermodynamical equilibrium. If the

decompression is too rapid, the expansion can induce visible consequences such as cracks

(Figure 3-3), blisters, voids and microstructures such as foams and thus may affect the

integrity of composite films particularly with an inorganic layer (Figure 3-4 and Figure

3-5) (BOYER, GROLIER 2005b; DEWIMILLE ET AL. 1993; RICHTER 2011). The occurring

phenomena can be summarized in the term ‘explosive decompression failure’ (XDF)

(BOYER, GROLIER 2005b; BRISCOE ET AL. 1994; BRISCOE ET AL. 1998). This phenomenon

is known from other technical areas (e.g. petroleum industry, BRISCOE, ZAKARIA (1992))

and is, e.g. intentionally used to produce polymeric foams (ARORA ET AL. 1998), whereas

it is unwanted in HPP, cf. (RICHTER 2010).

Figure 3-2 Plot of the equilibrium sorption amount of CO2 in PSU against pressure at 313K. Data taken from TANG ET AL. (2004b) and SADA ET AL. (1989). The wt% from SADA ET AL. was calculated with rPSU=1.24g/cm³.


99

Figure 3-3 Analysis of a local decompression failure by optical microscope, Raman microscopy and atomic force microscopy (AFM) (RICHTER 2011).

These occurring effects were detected by several authors: FAIRCLOUGH ET AL. reported

on small pits and sub-surface bubbles after HPP treatment (695 MPa; 86°C; 10min;

release 2s) of PP bags with an air headspace. Recently, RICHTER published Raman and

atomic force microscopy images of an aluminium foil delamination (Figure 3-4) and a

damaged SiOx coating (Figure 3-5). He traces both of these damages back on an explosive

decompression failure (500 MPa; 40°C; N2 and CO2 headspace) (RICHTER 2011).

Another group that reports on damages within polymeric packaging material that can be

allocated to explosive decompression failure is GÖTZ, WEISSER (2002). After high

pressure treatment (500 MPa) of PA/PE bags that were filled with CO2, bubbles arose

within the film, whereas O2 filled bags remained unaffected. This discrepancy may be

explained by the fact that in PE, the solubility of CO2 is higher than the solubility of O2

and a smaller amount of O2 will be dissolved in the polymer (VAN KREVELEN, NIJENHUIS

2009).


100

Also BULL ET AL. (2010) postulated that the delamination of some multilayer systems is

caused by explosive depressurization. In one case (600 MPa; 115°C; 5–10 min), the

expansion of the dissolved gas is reported to “push the laminated layers apart, leaving

behind a delaminated area”. The extent of delamination is correlated with the solubility

of the corresponding gases in the film layers. Multilayer systems with aluminium foil

showed less delamination than those without aluminium foil.

Figure 3-4 Microtome cut of a delamination (RICHTER 2011).

Figure 3-5 Damaged SiOx layer; Colour assignment in the right image: green, PET; blue, PE-LD; magenta, adhesive SiOx (RICHTER 2011).

The findings in the paper by FRADIN ET AL. (1998) from the earlier years of HPP research

also suggest an explosive decompression failure. They compared vacuum-sealed pouches

with pouches with a headspace of about 10%. In this paper, all multilayer specimens with

a gaseous headspace showed delamination and small bubbles, whereas a monolayer (PE-

LD) just showed a reduced transparency after high pressure treatment (200 MPa; 25°C;

15–45 min). The observed damages were independent of the duration of treatment.


101

Already at pressures under 50 MPa, explosive decompression failures can occur. So,

TANG ET AL. (2004b) found defects on PSU via a field emission scanning electron

microscope at their studies about sorption of supercritical carbon dioxide at pressures up

to 40MPa. There are some papers from the field of the petroleum industry dealing with

explosive decompression failures at lower pressures as well (BOYER ET AL. 2006; BOYER

ET AL. 2007; BOYER, GROLIER 2005b; HILIC ET AL. 2001).

Also, the circumscriptions occurrence of ‘voids’ (MASUDA ET AL. 1992) and ‘sporadic

blisters’ (KOUTCHMA ET AL. 2009) indicate explosive decompression failures, but there is

not enough information in these papers to allocate failures definitely. Different process

parameters and the damages caused by gas solubility in polymers after high pressure

treatment are summarized in Table 3-5.

Sorption-induced plasticization and crystallization. Next to explosive decompression

failures compressed gases lead to other effects that affect the integrity of polymeric

packaging. Some of these interactions of compressed gases with polymers have been

widely studied. First of all, CO2 is reported to acts as a plasticizer, especially in its

supercritical state (COOPER 2000). The most frequently used way to quantify

plasticization is the decrease of the glass transition temperature Tg of the polymers under

study. This was done in many studies (cf. Table 3-6) (CHIOU ET AL. 1985a; FRIED ET AL.

1989; HANDA ET AL. 1997; HOUDE ET AL. 1992, 1994; SANDERS 1988; WANG ET AL. 1982;

WISSINGER, PAULAITIS 1991; WONDERS, PAUL 1979). For a more detailed overview about

penetrant induced plasticization ISMAIL, LORNA (2002) see, e.g. the review of .

BOS ET AL. (1999) defined plasticization of glassy polymers as the increase in CO2

permeability as a function of fed pressure. In this paper, BOS ET AL. hypothesized “that all

polymers need a similar CO2 concentration to plasticize but require different pressures to

reach it.” This critical plasticization concentration was 71±14mg CO2/cm3 polymer for

the studied 11 polymers (inter alia PSU, polyethersulfone and polycarbonate).

The influence of pressure on Tg, however, is not always unidirectional. For polystyrene,

WANG ET AL. (1982) observed a minimum of Tg at a CO2 pressure of 20 MPa and a slight

increase of Tg at higher CO2 pressures (tested up to 105 MPa). Also the influence of

supercritical CO2 on the temperatures of crystallization Tc and melting Tm was tested.

KISHIMOTO, ISHII (2000) reported that compressed CO2 gas up to 9.4 MPa reduces Tm and

Tc of isotactic PP. These findings were confirmed by VARMA-NAIR ET AL. (2003).


102

Table 3-5 Solubility of headspace gases: damages.

Polymer Treatment Thickness Headspace Description Method Remark Reference

PA/PE 500 MPa 5 min

30/60 μm CO2 Bubbles, delamination

SEM (GÖTZ, WEISSER

2002) O2 No delamination

PP/PET 695 MPa 70 to 86°C 10 min

air Small pits, bubbles

SEM CLSM

Release 2s (FAIRCLOUGH, CONTI 2009)

PET/oPA6/PP*/**

600 MPa 115°C 5 to 10 min

12/25/80 μm

Less than 10 % of the packaging volume (air)

Large delamination

Visual detection

Independent of time and temperature

(BULL ET AL. 2010)

PET-AlOx/oPA6/PP 12/15/80 μm PET/oPA6/R-PP 12/15/80 μm

Delamination

PET/PVDC-MA/PP 12/50/70 μm PET/PA6/Al/R-PP 12/15/7/70 μm PET/PET-SiOx/PP 23/12/75 μm PET-SiOx/oPA/PP 12/15/75 μm PET-AlOx/oPA6/PP 12/15/70 μm PET-AlOx/oPA6/whitePP

12/15/100 μm

PET/Al/PP 12/7/70 μm Less delamination

PET/Al/PP 12/12/70 μm

PSU 20 to 40 MPa 40 to 60°C 24 h

0.8-1.2 mm CO2 Defects on surface FESEM (TANG ET AL. 2004b)

PE-LD

200 MPa 25°C 15 to 45 min

200 μm Less than 10 % of the packaging volume (air)

Reduced transparency

Visual

Independent of duration No damages at a vacuum of 25mbar

(FRADIN ET AL. 1998)

BB4L (PA6/PE)

Delamination Small bubbles

PA/PE 100 μm PA/SY 180 μm PET/PVDC/PE 65 μm PET/BOA/PE 62 μm n/a, not applicable; CLSM, confocal laser scanning microscopy; FESEM, field emission scanning microscope; SY, surlyn (copolymer of ethylene and methacrylic acid); PVDC, polyvinylidene chloride; BOA, benzyl-octyl-adipate. * o = oriented; ** PA6 = nylon


103

Table 3-6 Reduction of the glass transition temperature by solved CO2.

Polymer Tg (°C) ΔTg (°C)

CO2 pressure (MPa)

Remark Reference

untreated treated PVC 75 57 18

2.0

(CHIOU ET AL. 1985a)

PS 100 78 22 PC 148 97 51 PET 74 52 22 PMMA 105 67 38 Syndiotactic PS

97 65 32 3.5 * (HANDA ET AL. 1997)

PMMA 105 32.7-58.8

46.2-72.3

4.0 Dependent on the solved amount of CO2

(WISSINGER, PAULAITIS 1991)

PS 100 35 65 6.0 PES 235 135 100 3.5 (SANDERS 1988) PC 158 147 11

3.0

(FRIED ET AL. 1989)

PSU 197 179 18 PEI 219 213 6

PS 100 ~41 59 20.0 at 45 °C

Amount of plasticization is higher at lower ambient temperature*

(WANG ET AL. 1982) 100 46 54 92.0

PSU 286-291 260 26-31 2.0 PSU phenolphthaleine based

(HOUDE ET AL. 1992, 1994)

PC 142 133-134 8-9 0.8 * (WONDERS, PAUL 1979)

* Original data were represented graphically

The effect of plasticization, however, is not always a simple decrease in Tg. Because of

the increase in the mobility of the polymer chain segments (BRISCOE ET AL. 1998;

SCHULTZE ET AL. 1991), recrystallization may be induced, leading to an increase in

crystallinity and morphological damages (MA ET AL. 2004). The sorption-induced

increase in crystallinity was investigated in the first instance for CO2. Also there is only

information in the range of moderate pressures (cf. Table 3-7) (BECKMAN, PORTER 1987;

BRISCOE ET AL. 1998; CHIOU ET AL. 1985a; LAMBERT, PAULAITIS 1991; MIZOGUCHI ET

AL. 1987).

Even though these studies were conducted at moderate pressures (up to 50 MPa) in

comparison with HPP (~800 MPa) and longer treatment times (several hours to days vs.

several minutes at HPP), they show that HPP of packaging with modified atmosphere can

affect the morphology of the polymeric material.


104

Table 3-7 Sorption-induced increase in crystallinity

Polymer Crystallinity

(%) ΔCrystal-linity (%)

CO2 pressure (MPa)

Temp. (°C) Remark Reference

Un-treated

treated

PET 4.4 29.2 24.8 5.1 85 - (MIZOGUCHI

ET AL. 1987)

PET 8.0 29.0 21.0 4.1-6.1 35

Max. value independent of pressure Faster crystallization at higher pressures

(LAMBERT, PAULAITIS

1991)

PC 19.3 43.1 23.8 2.0

35 - (CHIOU ET

AL. 1985b) 19.3 48.7 29.4 3.5

PC A. 17.5 17.5

50.7 75 Bisphenol A

polycarbonate *

(BECKMAN, PORTER

1987) A. 22.0 22.0 88

PVDF

63.0 62.0 -1.0

30.0 42 and

80

Rods made from an extruded cylinder (BRISCOE ET

AL. 1998)

53.0 53.0 0.0 Parallelpiped made from injected sheet

* Original data were represented graphically A., Amorphous polymer; PVDF, polyvinylidene fluoride

3.3.2 Influence of non-gaseous substances

Delamination in multilayer systems without inorganic layer

In multilayer systems without inorganic layer, the reasons for the delamination are

interactions between the product and the tie layer or the pressure sensitivity of the tie

layer. For example, the former occurs in a PE/EVOH44/PE multilayer that is in contact

with oil and pressurized at 400 MPa for 30min at 20±2 or 60±2°C. In a parallel

experiment, the same multilayer shows no delamination in contact with water. A possible

explanation for this is an additional migration of low molecular compounds from the

adhesive into the oil (GALOTTO ET AL. 2010).

The latter was suggested by LAMBERT ET AL. (2000). In their study, six multilayer films

were analysed, which could be considered as identical except for the used adhesives. Only

one specimen (PA/adhesive/PE) showed delamination after treatment with 200–500 MPa


105

at 20°C for 30 min. Because all specifications and filling material except for the adhesive

were identical, LAMBERT ET AL. assumed that the adhesive is responsible for

delamination. However, the affected multilayer has an extraordinary thick adhesive layer

(20 µm) between PA and PE, and the specimen was sealed at a pressure of 80% below

atmospheric pressure. That means that a residual headspace gas exists, which can lead to

interactions (see Section on ‘Indirect effects’).

3.4 Conclusion

High pressure-induced damages in polymeric packaging materials can be assigned to

different mechanisms: effects that are directly caused by the high pressure treatment

(direct effects) and effects that are generated by the compression of other substances in

the package (indirect effects). To achieve a maximum of functionality in high pressure-

treated packaging, those mechanisms have to be identified.

Direct effects

During high pressure treatment, the permeation coefficient for some polymer/substance

combinations shows to decrease reversibly, mainly because of a decrease in the diffusion

coefficient. In the common free-volume model of permeation, this can be directly

assigned to a pressure-induced reduction of the free volume. Although a qualitative

correlation between polymer morphology and permeation has so far not been established,

the temporal and reversible decrease in permeability seems to be a general phenomenon,

which, besides, does not negatively affect the functionality of the packaging.

Another direct effects of high pressure treatment on polymers are changes in crystallinity.

Because of several different measurement methods and the application of different blends

and parameters, the various published papers do not show homogeneous findings.

Permanent effects of HPP are directly reflected by changes in the morphology in the bulk

of polymers or at the interface of polymeric and inorganic layers. First, changes in

crystallinity have been observed by different authors, but the findings do not show a clear

trend. Second, delamination or other damages have been found by many authors for

multilayer systems. These effects have a clearly negative effect on the functionality of the

whole package.


106

In the case of multilayer systems that include inorganic layers, both direct damages of the

inorganic layers and delamination at the interface between inorganic and polymeric layers

are frequently observed. Both mechanisms can be assigned to the discontinuity of the

mechanical properties across the interface, which a priori cannot be totally avoided. It is

certainly a good strategy to improve the adhesion between the layers, but, as a general

rule, polymeric multilayer systems without inorganic layers will be superior.

Indirect effects

Indirect effects represent all influences of compressed substances inside the package.

Primarily, these effects are caused by compressed gases. They will first generate thermal

effects, ranging from, e.g. localized increase of crystallinity up to catastrophic failures

such as melting of the sealing layer.

Another potential effect can be summarized as solution effects. There, the higher

concentration of gases in the polymeric matrix at high pressure can lead to a plasticization

of polymers, followed by structural changes or an extraction of constituents from the

polymer enabled by the supercritical state. In addition, (explosive) decompression failures

may occur because of decompression, which leads to an oversaturation of the gas in the

polymer. These failures arise in the form of cracks, blistering or other microstructures and

turn up to concentrate in weak zones of the materials and, again, in interfacial zones.

As a final conclusion, it can be stated that the impacts of HPP on barrier properties and

visual appearance as the most important properties of plastic packaging are treated

phenomenologically at best. Many questions concerning physicochemical processes in

the polymeric materials still remain unanswered: How does the morphology change upon

the action of high pressure, especially in combination with supercritical gases? Does high

pressure affect gas sorption and permeation in the same way as already reported for aroma

substances? As the answers to these questions are crucial for a wider application of HPP,

many work still has to be done in this area.


107


AHMED, A.; AHMED, N.; SALMAN, A. (2005): Critical issues in packaged food business.

In British Food Journal 107 (10), pp. 760–780.

AHMED, J.; VARSHNEY, S. K.; ZHANG, J.-X.; RAMASWAMY, H. S. (2009): Effect of high

pressure treatment on thermal properties of polylactides. In Journal of Food

Engineering 93 (3), pp. 308–312.

ANGELIS, M. G. de; SARTI, G. C. (2011): Solubility of gases and liquids in glassy

polymers. In Annual review of chemical and biomolecular engineering 2, pp. 97–

120. DOI: 10.1146/annurev-chembioeng-061010-114247.

ARORA, K. A.; LESSER, A. J.; MCCARTHY, T. J. (1998): Preparation and

characterization of microcellular polystyrene foams processed in supercritical carbon

dioxide. In Macromolecules 31 (14), pp. 4614–4620. DOI: 10.1021/ma971811z.

AYVAZ, H.; SCHIRMER, S.; PARULEKAR, Y.; BALASUBRAMANIAM, V. M.; SOMERVILLE,

J. A.; DARYAEI, H. (2012): Influence of selected packaging materials on some

quality aspects of pressure-assisted thermally processed carrots during storage. In

LWT-Food Science and Technology 46 (2), pp. 437–447.

BALASUBRAMANIAM, V. M.; FARKAS, D. F. (2008): High-pressure food processing. In

Food Science and Technology International 14 (5), pp. 413–418. DOI:

10.1177/1082013208098812.

BALASUBRAMANIAN, S.; BALASUBRAMANIAM, V. M. (2003): Compression heating

influence of pressure transmitting fluids on bacteria inactivation during high pressure

processing. In Food Research International 36 (7), pp. 661–668. DOI:

10.1016/S0963-9969(03)00014-0.

BARBOSA-CÁNOVAS, G. V.; JULIANO, P. (2008): Food sterilization by combining high

pressure and thermal energy. In Gustavo F. Gutiérrez-López (Ed.): Food engineering.

Integrated approaches. New York: Springer (Food Engineering Series), pp. 9–46.

BARBOSA-CÁNOVAS, G. V.; RODRIGUEZ, J. J. (2005): Thermodynamic aspects of high

hydrostatic pressure food processing. In Gustavo V. Barbosa-Cánovas, M. S. Tapia,

C. M. Pilar (Eds.): Novel food processing technologies. Boca Raton: CRC Press,

pp. 183–205.


108

BARRER, R. M. (1951): Diffusion in and through solids. Cambridge: University Press

(The Cambridge series of physical chemistry).

BECKMAN, E.; PORTER, R. S. (1987): Crystallization of bisphenol a polycarbonate

induced by supercritical carbon dioxide. In J. Polym. Sci. B Polym. Phys. 25 (7),

pp. 1511–1517. DOI: 10.1002/polb.1987.090250713.

BERET, S.; PRAUSNITZ, J. M. (1974): Densities of amorphous polymers at high

pressures. In Journal of Applied Polymer Science 18 (12), pp. 3779–3789.

BERMÚDEZ-AGUIRRE, D.; BARBOSA-CÁNOVAS, G. V. (2011): An update on high

hydrostatic pressure, from the laboratory to industrial applications. In Food Eng. Rev.

3 (1), pp. 44–61. DOI: 10.1007/s12393-010-9030-4.

BOS, A.; PÜNT, I. G. M.; WESSLING, M.; STRATHMANN, H. (1999): CO2-induced

plasticization phenomena in glassy polymers. In Journal of Membrane Science 155

(1), pp. 67–78. DOI: 10.1016/S0376-7388(98)00299-3.

BOYER, S. A. E.; GROLIER, J.-P. E. (2005): Simultaneous measurement of the

concentration of a supercritical gas absorbed in a polymer and of the concomitant

change in volume of the polymer. The coupled VW-pVT technique revisited. In

Polymer 46 (11), pp. 3737–3747.




10.1002/app.25085.

BOYER, S. A. E.; RANDZIO, S. L.; GROLIER, J.-P. E. (2006): Thermal expansion of

polymers submitted to supercritical CO2 as a function of pressure. In Journal of

Polymer Science Part B: Polymer Physics 44 (1), pp. 185–194. DOI:

10.1002/polb.20674.

BRISCOE, B. J.; LORGE, O.; WAJS, A.; DANG, P. (1998): Carbon dioxide–

poly(vinylidene fluoride) interactions at high pressure. In Journal of Polymer Science

Part B: Polymer Physics 36 (13), pp. 2435–2447. DOI: 10.1002/(SICI)1099-

0488(19980930)36:13.


109

BRISCOE, B. J.; SAVVAS, T.; KELLY, C. T. (1994): “Explosive decompression failure” of

rubbers: A review of the origins of pneumatic stress induced rupture in elastomers. In

Rubber Chemistry and Technology 67 (3), pp. 384–416. DOI: 10.5254/1.3538683.

BRISCOE, B. J.; ZAKARIA, S. (1992): Sorption and dilation of silicone elastomer

composites at high gas pressures: The role of interfacial quality. In J. Polym. Sci. B

Polym. Phys. 30 (9), pp. 959–969. DOI: 10.1002/polb.1992.090300904.

BRUNNER, G. (2010): Applications of supercritical fluids. In Annual review of chemical

and biomolecular engineering 1, pp. 321–342. DOI: 10.1146/annurev-chembioeng-

073009-101311.





BUZRUL, S.; ALPAS, H.; LARGETEAU, A.; BOZOGLU, F.; DEMAZEAU, G. (2008):

Compression heating of selected pressure transmitting fluids and liquid foods during

high hydrostatic pressure treatment. In Journal of Food Engineering 85 (3), pp. 466–

472. DOI: 10.1016/j.jfoodeng.2007.08.014.

CAMPION, R. P.; MORGAN, G. J. (1992): High pressure permeation and diffusion of

gases in polymers of different structures. In Plastics, Rubber and Composites

Processing and Applications 17 (1), pp. 51–58.

CANER, C.; HARTE, B. R. (2005): Effect of high-pressure processing on the migration of

antioxidant Irganox 1076 from polypropylene film into a food simulant. In J. Sci.

Food Agric. 85 (1), pp. 39–46. DOI: 10.1002/jsfa.1935.

CANER, C.; HERNANDEZ, R. J.; HARTE, B. R. (2004a): High-pressure processing effects

on the mechanical, barrier and mass transfer properties of food packaging flexible

structures: a critical review. In Packaging Technology & Science 17 (1), pp. 23–29.

DOI: 10.1002/pts.635.

CANER, C.; HERNANDEZ, R. J.; PASCALL, M.; BALASUBRAMANIAM, V. M.; HARTE, B.

R. (2004b): The effect of high-pressure food processing on the sorption behaviour of


110


153. DOI: 10.1002/pts.650.

CANER, C.; HERNANDEZ, R. J.; PASCALL, M. A.; RIEMER, J. (2003): The use of

mechanical analyses, scanning electron microscopy and ultrasonic imaging to study

the effects of high-pressure processing on multilayer films. In J. Sci. Food Agric. 83

(11), pp. 1095–1103. DOI: 10.1002/jsfa.1437.

CHANDRA, P.; KOROS, W. J. (2009): Sorption of lower alcohols in poly(ethylene

terephthalate). In Polymer 50 (17), pp. 4241–4249. DOI:

10.1016/j.polymer.2009.06.066.

CHIOU, J. S.; BARLOW, J. W.; PAUL, D. R. (1985a): Plasticization of glassy polymers by

CO2. In J. Appl. Polym. Sci. 30 (6), pp. 2633–2642. DOI:

10.1002/app.1985.070300626.

CHIOU, J. S.; BARLOW, J. W.; PAUL, D. R. (1985b): Polymer crystallization induced by

sorption of CO2 gas. In Journal of Applied Polymer Science 30 (9), pp. 3911–3924.

DOI: 10.1002/app.1985.070300929.

COOPER, A. I. (2000): Polymer synthesis and processing using supercritical carbon

dioxide. In Journal of Materials Chemistry 10 (2), pp. 207–234.

COTTON, N. J.; BARTLE, K. D.; CLIFFORD, A. A.; ASHRAF, S.; MOULDER, R.; DOWLE,

C. J. (1991): Analysis of low molecular weight constituents of polypropylene and

other polymeric materials using on-line SFE-SFC. In J. High Resol. Chromatogr. 14

(3), pp. 164–168. DOI: 10.1002/jhrc.1240140305.

DENYS, S.; LUDIKHUYZE, L. R.; VAN LOEY, A. M.; HENDRICKX, M. E. (2000): Modeling

conductive heat transfer and process uniformity during batch high-pressure

processing of foods. In Biotechnology progress 16 (1), pp. 92–101. DOI:

10.1021/bp990123r.

DEWIMILLE, B.; MARTIN, J.; JARRIN, J. (1993): Behaviour of thermoplastic polymers

during explosive decompressions in a petroleum environment. In Le Journal de

Physique IV 3 (C7), p. 1559.

DHAWAN, S.; BARBOSA-CÁNOVAS, G. V.; TANG, J.; SABLANI, S. S. (2011): Oxygen

barrier and enthalpy of melting of multilayer EVOH films after pressure-assisted


111

thermal processing and during storage. In Journal of Applied Polymer Science 122

(3), pp. 1538–1545. DOI: 10.1002/app.34267.

DILLER, D. E. (1983): Measurements of the viscosity of compressed gaseous and liquid

nitrogen. In Physica A: Statistical Mechanics and its Applications 119 (1-2), pp. 92–

100. DOI: 10.1016/0378-4371(83)90149-8.

DOBIÁS, J.; VOLDRICH, M.; MAREK, M.; CHUDÁCKOVÁ, K. (2004): Changes of

properties of polymer packaging films during high pressure treatment. Food

Processing and Technology - Selected Papers from the 15th CHISA Congress. In

Journal of Food Engineering 61 (4), pp. 545–549.

EMSLEY, J. (1989): The elements. Oxford, New York: Clarendon Press; Oxford

University Press.

ESLAMI, H.; MÜLLER-PLATHE, F. (2007): Molecular dynamics simulation of sorption of

gases in polystyrene. In Macromolecules 40 (17), pp. 6413–6421.



pp. 303–310. DOI: 10.1002/pts.858.

FERNANDEZ, J. O.; SWALLOWE, G. M. (2000): Crystallisation of PET with strain, strain

rate and temperature. In Journal of Materials Science 35 (17), pp. 4405–4414. DOI:

10.1023/A:1004817325954.




FRIED, J. R.; LIU, H.-C.; ZHANG, C. (1989): Effect of sorbed carbon dioxide on the

dynamic mechanical properties of glassy polymers. In J. Polym. Sci. C Polym. Lett.

27 (10), pp. 385–392. DOI: 10.1002/pol.1989.140271007.

GALIĆ, K.; ŠČETAR, M.; KUREK, M. (2011): The benefits of processing and packaging.

In Trends in Food Science & Technology 22 (2-3), pp. 127–137. DOI:

10.1016/j.tifs.2010.04.001.

GALOTTO, M. J.; ULLOA, P. A.; ESCOBAR, R.; GUARDA, A.; GAVARA, R.; MILTZ, J.

(2010): Effect of high-pressure food processing on the mass transfer properties of


112


266. DOI: 10.1002/pts.893.

GALOTTO, M. J.; ULLOA, P. A.; GUARDA, A.; GAVARA, R.; MILTZ, J. (2009): Effect of

High-Pressure Food Processing on the Physical Properties of Synthetic and

Biopolymer Films. In J Food Sci 74 (6), pp. E304-E311. DOI: 10.1111/j.1750-

3841.2009.01212.x.





GORLIER, E.; HAUDIN, J. M.; BILLON, N. (2001): Strain-induced crystallisation in bulk

amorphous PET under uni-axial loading. In Polymer 42 (23), pp. 9541–9549.

GÖTZ, J.; WEISSER, H. (2002): Permeation of aroma compounds through plastic films

under high pressure: in-situ measuring method. In Innovative Food Science &


GRUVER, V.; SHOWERS, D.; KAO, M.; KLEBANOV, L. (2000): The determination of PET

crystallinity by different analytical techniques. In Society of Plastics Engineers (Ed.):

Conference Proceesings ANTEC 2000. Plastics - the magical solution 2. Lancester,

pp. 1636–1640.

GUILLARD, V.; MAURICIO-IGLESIAS, M.; GONTARD, N. (2010): Effect of novel food

processing methods on packaging: Structure, composition, and migration properties.

Critical Reviews in Food Science and Nutrition. In Critical reviews in Food Science

and Nutrition 50 (10), pp. 969–988. DOI: 10.1080/10408390903001768.

GUPTA, R.; BALASUBRAMANIAM, V. M. (2011): High-pressure processing of fluid

foods. In Patrick J. Cullen, Brijesh K. Tiwari (Eds.): Novel thermal and non-thermal

technologies for fluid foods. San Diego: Academic Press (Food Science and

Technology), pp. 109–133.

HANDA, Y. P.; WONG, B.; ZHANG, Z.; KUMAR, V.; EDDY, S.; KEHMANI, K. (1999):

Some thermodynamic and kinetic properties of the system PETG-CO2, and


113

morphological characteristics of the CO2-blown PETG foams. In Polymer

engineering and science 39 (1), pp. 55–61.

HANDA, Y. P.; ZHANG, Z.; WONG, B. (1997): Effect of compressed CO2 on phase

transitions and polymorphism in syndiotactic polystyrene. In Macromolecules 30

(26), pp. 8499–8504. DOI: 10.1021/ma9712209.

HARTMANN, C.; DELGADO, A.; SZYMCZYK, J. (2003): Convective and diffusive

transport effects in a high pressure induced inactivation process of packed food. In

Journal of Food Engineering 59 (1), pp. 33–44.

HILIC, S.; BOYER, S. A. E.; PÁDUA, A. A. H.; GROLIER, J.-P. E. (2001): Simultaneous

measurement of the solubility of nitrogen and carbon dioxide in polystyrene and of

the associated polymer swelling. In Journal of Polymer Science Part B: Polymer

Physics 39 (17), pp. 2063–2070. DOI: 10.1002/polb.1181.

HOUDE, A. Y.; KULKARNI, S. S.; KULKARNI, M. G. (1992): Permeation and

plasticization behavior of glassy polymers - a WAXD interpretation. In Journal of

Membrane Science 71 (1-2), pp. 117–128. DOI: 10.1016/0376-7388(92)85011-7.

HOUDE, A. Y.; KULKARNI, S. S.; KULKARNI, M. G. (1994): Sorption, transport, and

history effects in phenolphthalein-based polysulfone. In Journal of Membrane

Science 95 (2), pp. 147–160. DOI: 10.1016/0376-7388(94)00106-5.

ISMAIL, A. F.; LORNA, W. (2002): Penetrant-induced plasticization phenomenon in

glassy polymers for gas separation membrane. In Separation and Purification

Technology 27 (3), pp. 173–194. DOI: 10.1016/S1383-5866(01)00211-8.

JORDAN, S. M.; KOROS, W. J.; BEASLEY, J. K. (1989): Characterization of CO2-induced

conditioning of polycarbonate films using penetrants with different solubilities. In

Journal of Membrane Science 43 (1), pp. 103–120. DOI: 10.1016/S0376-

7388(00)82356-X.



Reviews 2 (4), pp. 274–297.


114

JUNG, S.; SAMSON, C. T.; LAMBALLERIE, M. de (2011): Chapter 6: High hydrostatic

pressure food processing. In Andrew Proctor (Ed.): Alternatives to conventional food

processing. Cambridge: RSC (RSC green chemistry, no. 10), pp. 254–306.

KISHIMOTO, Y.; ISHII, R. (2000): Differential scanning calorimetry of isotactic

polypropene at high CO2 pressures. In Polymer 41 (9), pp. 3483–3485. DOI:

10.1016/S0032-3861(99)00698-9.

KLOPFFER, M.-H.; FLACONNÈCHE, B. (2001): Transport properties of gases in polymers.

Bibliographic Review. In Oil Gas Sci. Technol. – Rev. IFP Energies nouvelles 56

(3), pp. 223–244.

KNOERZER, K.; BUCKOW, R.; VERSTEEG, C. (2010): Adiabatic compression heating

coefficients for high-pressure processing–a study of some insulating polymer

materials. In Journal of Food Engineering 98 (1), pp. 110–119.

KOUTCHMA, T.; SONG, Y.; SETIKAITE, I.; JULIANO, P.; BARBOSA-CÁNOVAS, G. V.;

DUNNE, C. P.; PATAZCA, E. (2009): Packaging evaluation for high-pressure high-

temperature sterilization of shelf-stable foods. In Journal of Food Process

Engineering, pp. no-no. DOI: 10.1111/j.1745-4530.2008.00328.x.

KOVARSKII, A. L. (1994): High-pressure chemistry and physics of polymers. Boca

Raton: CRC Press.

KÜBEL, J.; LUDWIG, H.; MARX, H.; TAUSCHER, B. (1996): Diffusion of Aroma

Compounds into Packaging Films under High Pressure. In Packaging Technology &

Science 9 (3), pp. 143–152. DOI: 10.1002/(SICI)1099-1522(199605)9:3.

LAMBERT, S. M.; PAULAITIS, M. E. (1991): Crystallization of poly(ethylene

terephthalate) induced by carbon dioxide sorption at elevated pressures. In The

Journal of Supercritical Fluids 4 (1), pp. 15–23. DOI: 10.1016/0896-8446(91)90026-

3.

LAMBERT, Y.; DEMAZEAU, G.; LARGETEAU, A.; BOUVIER, J. M.; LABORDE-CROUBIT,

S.; CABANNES, M. (2000): Packaging for high-pressure treatments in the food

industry. In Packaging Technology & Science 13 (2), pp. 63–71.

LARGETEAU, A. (2010): Evaluation of films for packaging applications in high pressure

processing. In Journal of Physics: Conference Series 215 (1), pp. 1–11.


115

LAX, E.; ANS, J. d.; BLACHNIK, R. (1998): Taschenbuch fur Chemiker und Physiker.

Elemente, anorganische Verbindungen und Materialien, Minerale. Band 3. Berlin

Heidelberg: Springer DE (Elemente, anorganische Verbindungen und Materialien,

Minerale).

LETERRIER, Y. (2003): Durability of nanosized oxygen-barrier coatings on polymers. In

Progress in Materials Science 48 (1), pp. 1–55. DOI: 10.1016/S0079-

6425(02)00002-6.

LETERRIER, Y.; ANDERSONS, J.; PITTON, Y.; MANSON, J.-A. E. (1997): Adhesion of

silicon oxide layers on poly (ethylene terephthalate). II. Effect of coating thickness

on adhesive and cohesive strengths. In J. Polym. Sci. B Polym. Phys. 35 (9),

pp. 1463–1472. DOI: 10.1002/(SICI)1099-0488(19970715)35:9<1463::AID-

POLB16>3.0.CO;2-4.

LÓPEZ-RUBIO, A.; LAGARÓN, J. M.; GIMENEZ, E.; CAVA, D.; HERNÁNDEZ-MUÑOZ, P.;

YAMAMOTO, T.; GAVARA, R. (2003): Morphological alterations induced by

temperature and humidity in ethylene-vinyl alcohol copolymers. In Macromolecules

36 (25), pp. 9467–9476. DOI: 10.1021/ma035346j.

LÓPEZ-RUBIO, A.; LAGARÓN, J. M.; HERNÁNDEZ-MUÑOZ, P.; ALMENAR, E.; CATALÁ,

R.; GAVARA, R.; PASCALL, M. A. (2005): Effect of high pressure treatments on the

properties of EVOH-based food packaging materials. In Innovative Food Science &


LUDWIG, H.; MARX, H.; TAUSCHER, B. (1994): Interaction of food packaging material

and selected food components under high pressure. In High Pressure Research 12 (4-

6), pp. 251–254. DOI: 10.1080/08957959408201665.

MA, W.; YU, J.; HE, J. (2004): Direct formation of γ form crystal of syndiotactic

polystyrene from amorphous state in supercritical CO2. In Macromolecules 37 (18),

pp. 6912–6917. DOI: 10.1021/ma0491715.







116

MAURICIO-IGLESIAS, M.; JANSANA, S.; PEYRON, S.; GONTARD, N.; GUILLARD, V.

(2010): Effect of high-pressure/temperature (HP/T) treatments of in-package food on

additive migration from conventional and bio-sourced materials. In Food Additives &

Contaminants: Part A: Chemistry, Analysis, Control, Exposure & Risk Assessment

27 (1), pp. 118–127. DOI: 10.1080/19440040903268054.

MAURICIO-IGLESIAS, M.; PEYRON, S.; CHALIER, P.; GONTARD, N. (2011): Scalping of

four aroma compounds by one common (LDPE) and one biosourced (PLA)

packaging materials during high pressure treatments. In Journal of Food Engineering

102 (1), pp. 9–15.

MERTENS, B. (1993): Packaging aspects of high-pressure food processing technology.

In Packaging Technology & Science 6 (1), pp. 31–36. DOI:

10.1002/pts.2770060106.

MIN, S.; ZHANG, Q. H. (2007): Packaging for high-pressure processing, irradiation, and

pulsed electric field processing. In Jung H. Han (Ed.): Packaging for nonthermal

processing of food. 1st. Iowa: Blackwell Pub.; IFT Press, pp. 67–86.

MIZOGUCHI, K.; HIROSE, T.; NAITO, Y.; KAMIYA, Y. (1987): CO2-induced

crystallization of poly(ethylene terephthalate). In Polymer 28 (8), pp. 1298–1302.

DOI: 10.1016/0032-3861(87)90441-1.

MORRIS, C.; BRODY, A. L.; WICKER, L. (2007): Non-thermal food

processing/preservation technologies: a review with packaging implications. In

Packaging Technology & Science 20 (4), pp. 275–286. DOI: 10.1002/pts.789.

NORTON, T.; SUN, D.-W. (2008): Recent advances in the use of high pressure as an

effective processing technique in the food industry. In Food and Bioprocess

Technology 1 (1), pp. 2–34. DOI: 10.1007/s11947-007-0007-0.





519.


117

PATAZCA, E.; KOUTCHMA, T.; BALASUBRAMANIAM, V. M. (2007): Quasi-adiabatic

temperature increase during high pressure processing of selected foods. In Journal of

Food Engineering 80 (1), pp. 199–205. DOI: 10.1016/j.jfoodeng.2006.05.014.

PATERSON, R.; YAMPOL’SKII, Y.; FOGG, P. G. T.; BOKAREV, A.; BONDAR, V.; ILINICH,

O.; SHISHATSKII, S. (1999): IUPAC-NIST Solubility Data Series 70. Solubility of

Gases in Glassy Polymers. In J. Phys. Chem. Ref. Data 28 (5), p. 1255. DOI:

10.1063/1.556050.

RASANAYAGAM, V.; BALASUBRAMANIAM, V. M.; TING, E.; SIZER, C. E.; BUSH, C.;

ANDERSON, C. (2003): Compression heating of selected fatty food materials during

high-pressure processing. In J Food Sci 68 (1), pp. 254–259. DOI: 10.1111/j.1365-

2621.2003.tb14148.x.

RASTOGI, N. K.; RAGHAVARAO, K.; BALASUBRAMANIAM, V. M.; NIRANJAN, K.;

KNORR, D. (2007): Opportunities and challenges in high pressure processing of

foods. In Critical reviews in Food Science and Nutrition 47 (1), pp. 69–112.

Commission Regulation (EU) No 10/2011, 1/14/2011: Regulations on plastic materials

and articles intended to come into contact with food.

RICHTER, T. (2010): Machine and method for packaging and high pressure treatment of

products. App. no. US 2011/0067355 A1.





566. DOI: 10.1080/08957959.2010.531722.

RIVAS-CAÑEDO, A.; FERNANDEZ-GARCIA, E.; NUNEZ, M. (2009): Volatile compounds

in fresh meats subjected to high pressure processing: Effect of the packaging

material. In Meat science 81 (2), pp. 321–328. DOI: 10.1016/j.meatsci.2008.08.008.

RIVAS-CAÑEDO, A.; JUEZ-OJEDA, C.; NUÑEZ, M.; FERNÁNDEZ-GARCÍA, E. (2011):

Volatile compounds in ground beef subjected to high pressure processing. A

comparison of dynamic headspace and solid-phase microextraction. In Food

Chemistry 124 (3), pp. 1201–1207. DOI: 10.1016/j.foodchem.2010.07.045.


118

RODERT, H. M. (1982): The thermal conductivity of oxygen. In Measurements of the

Octanol/Water Partition Coefficient by Chromatographic Methods 87 (4), p. 279.

RUNT, J.; KANCHANASOPA, M. (2004): Crystallinity determination. In : Encyclopedia of

polymer science and technology. Hoboken, NJ: Wiley Interscience.

SADA, E.; KUMAZAWA, H.; XU, P. (1989): Permeation of carbon dioxide through

homogeneous and asymmetric polysulfone membranes. In J. Polym. Sci. B Polym.

Phys. 27 (4), pp. 919–927. DOI: 10.1002/polb.1989.090270415.

SADA, E.; KUMAZAWA, H.; XU, P.; NISHIGAKI, M. (1988): Mechanism of gas

permeation through glassy polymer films. In Journal of Membrane Science 37 (2),

pp. 165–179. DOI: 10.1016/S0376-7388(00)83070-7.

SAN MARTÍN, M. F.; BARBOSA-CÁNOVAS, G. V.; SWANSON, B. G. (2002): Food

processing by high hydrostatic pressure. In Critical reviews in Food Science and

Nutrition 42 (6), pp. 627–645. DOI: 10.1080/20024091054274.

SANDERS, E. S. (1988): Penetrant-induced plasticization and gas permeation in glassy

polymers. In Journal of Membrane Science 37 (1), pp. 63–80. DOI: 10.1016/S0376-

7388(00)85069-3.

SANSONE, L.; ALDI, A.; MUSTO, P.; DI MAIO, E.; AMENDOLA, E.; MENSITIERI, G.

(2012): Assessing the suitability of polylactic acid flexible films for high pressure

pasteurization and sterilization of packaged foodstuff. In Journal of Food

Engineering 111 (1), pp. 34–45. DOI: 10.1016/j.jfoodeng.2012.01.034.

SCHAUWECKER, A.; BALASUBRAMANIAM, V. M.; SADLER, G.; PASCALL, M. A.;

ADHIKARI, C. (2002): Influence of high-pressure processing on selected polymeric

materials and on the migration of a pressure-transmitting fluid. In Packaging


SCHEICHL, R.; KLOPFFER, M.-H.; BENJELLOUN-DABAGHI, Z.; FLACONNÈCHE, B. (2005):

Permeation of gases in polymers: parameter identification and nonlinear regression

analysis. In Journal of Membrane Science 254 (1-2), pp. 275–293.

SCHICK, C. (2009): Differential scanning calorimetry (DSC) of semicrystalline

polymers. In Analytical and Bioanalytical Chemistry 395 (6), pp. 1589–1611. DOI:

10.1007/s00216-009-3169-y.


119

SCHMERDER, A.; RICHTER, T.; LANGOWSKI, H.-C.; LUDWIG, H. (2005): Effect of high

hydrostatic pressure on the barrier properties of polyamide-6 films. In Braz J Med

Biol Res 38 (8), pp. 1279–1283. Available online at doi:/S0100-

879X2005000800018.

SCHULTZE, J. D.; ENGELMANN, I. A. D.; BOEHNING, M.; SPRINGER, J. (1991): Influence

of sorbed carbon dioxide on transition temperatures of poly(p-phenylene sulphide).

In Polym. Adv. Technol. 2 (3), pp. 123–126. DOI: 10.1002/pat.1991.220020303.

SOLMS, N. von; MICHELSEN, M. L.; KONTOGEORGIS, G. M. (2005): Prediction and

correlation of high-pressure gas solubility in polymers with simplified PC-SAFT. In

Ind. Eng. Chem. Res. 44 (9), pp. 3330–3335.

SPAN, R.; WAGNER, W. (1996): A new equation of state for carbon dioxide covering the

fluid region from the triple-point temperature to 1100 K at pressures up to 800 MPa.

In J. Phys. Chem. Ref. Data 25 (6), p. 1509. DOI: 10.1063/1.555991.

TANG, M.; HUANG, Y.-C.; CHEN, Y.-P. (2004): Sorption and diffusion of supercritical

carbon dioxide into polysulfone. In J. Appl. Polym. Sci. 94 (2), pp. 474–482. DOI:

10.1002/app.20895.

TANG, Q.; YANG, B.; ZHAO, Y.; ZHAO, L. (2007): Sorption and diffusion of

sub/supercritical carbon dioxide in poly (methyl methacrylate). In Journal of

Macromolecular Science, Part B: Physics 46 (2), pp. 275–284.

The European Commission (2/23/1990): Commission Directive 90/128/EEC of 23

February 1990 relating to plastic materials and articles intended to come into contact

with foodstuffs. 90/128/EEC. In Official Journal of the European Communities (33

(L 75)), pp. 19–40.

TING, E.; BALASUBRAMANIAM, V. M. (2002): Determining thermal effects in high-

pressure processing. In Food Technology 56 (2), pp. 31–35.

TOEPFL, S.; MATHYS, A.; HEINZ, V.; KNORR, D. (2006): Review: Potential of high

hydrostatic pressure and pulsed electric fields for energy efficient and

environmentally friendly food processing. In Food Reviews International 22 (4),

pp. 405–423. DOI: 10.1080/87559120600865164.


120

TORRES, J. A.; VELAZQUEZ, G. (2005): Commercial opportunities and research

challenges in the high pressure processing of foods. IV Iberoamerican Congress of

Food Engineering (CIBIA IV). In Journal of Food Engineering 67 (1-2), pp. 95–112.

VAN KREVELEN, D. W.; NIJENHUIS, K. t. (2009): Properties of polymers. Their

correlation with chemical structure; their numerical estimation and prediction from

additive group contributions. 4th, completely rev. ed. Amsterdam, Boston: Elsevier.

VARMA-NAIR, M.; HANDA, P. Y.; MEHTA, A. K.; AGARWAL, P. (2003): Effect of

compressed CO2 on crystallization and melting behavior of isotactic polypropylene.

In Thermochimica Acta 396 (1-2), pp. 57–65. DOI: 10.1016/S0040-6031(02)00516-

6.

VENEMA, A.; VAN DE VEN, H. J. F. M.; DAVID, F.; SANDRA, P. (1993): Supercritical

fluid extraction of nylon 6: An investigation into the factors affecting the efficiency

of extraction of caprolactam and its oligomers. In J. High Resol. Chromatogr. 16 (9),

pp. 522–524. DOI: 10.1002/jhrc.1240160904.

VIETH, W. R. (1991): Diffusion in and through polymers. Principles and applications.

Munich, Vienna, New York, Barcelona: Hanser Publishers; Distributed in the USA

and in Canada by Oxford University Press.

VRENTAS, J. S.; DUDA, J. L. (1977): Diffusion in polymer—solvent systems. I.

Reexamination of the free‐volume theory. In Journal of Polymer Science: Polymer

Physics Edition 15 (3), pp. 403–416.

WANG, W.-C. V.; KRAMER, E. J.; SACHSE, W. H. (1982): Effects of high-pressure CO2

on the glass transition temperature and mechanical properties of polystyrene. In J.

Polym. Sci. Polym. Phys. Ed. 20 (8), pp. 1371–1384. DOI:

10.1002/pol.1982.180200804.

WISSINGER, R. G.; PAULAITIS, M. E. (1991): Glass transitions in polymer / CO2

mixtures at elevated pressures. In J. Polym. Sci. B Polym. Phys. 29 (5), pp. 631–633.

DOI: 10.1002/polb.1991.090290513.

WONDERS, A. G.; PAUL, D. R. (1979): Effect of CO2 exposure history on sorption and

transport in polycarbonate. In Journal of Membrane Science 5, pp. 63–75. DOI:

10.1016/S0376-7388(00)80438-X.


121

YOO, S.; LEE, J.; HOLLOMAN, C.; PASCALL, M. A. (2009): The effect of high pressure

processing on the morphology of polyethylene films tested by differential scanning

calorimetry and X-ray diffraction and its influence on the permeability of the

polymer. In J. Appl. Polym. Sci. 112 (1), pp. 107–113. DOI: 10.1002/app.29401.

ZENG, C.; HAN, X.; LEE, L. J.; KOELLING, K. W.; TOMASKO, D. L. (2003): Polymer–

clay nanocomposite foams prepared using carbon dioxide. In Adv. Mater. 15 (20),

pp. 1743–1747. DOI: 10.1002/adma.200305065.

ZIELINSKI, J. M.; DUDA, J. L. (1992): Predicting polymer/solvent diffusion coefficients

using free-volume theory. In AIChE J. 38 (3), pp. 405–415. DOI:

10.1002/aic.690380309.

122

PUBLICATION III CHAPTER 4

123

4 In-situ measurement of oxygen concentration under high

pressure and the application to oxygen permeation through

polymer films

Up until now, gas permeation through polymers under high pressure has not been able to

be measured continuously. The combination of a special high pressure cell and a

commercially available fluorescence-based oxygen measurement system allows in-situ

monitoring of oxygen permeation through a polymer sample under pressure in an aqueous

environment. The principle of the oxygen sensor is based on dynamic fluorescence

quenching and measurement of the fluorescence decay time. It was observed that the

decay time increases non-linearly with the applied pressure and hence the displayed

oxygen concentration has to be corrected. This deviation between the measured and the

real concentration depends not only on the pressure but also on the absolute oxygen

concentration in the water. To obtain a calibration curve, tests were performed in the

pressure range between 1 and 2,000 bars and initial oxygen concentrations in the range

between 40 and 280 µmol/L. The polynomial calibration curve was of the fourth order,

describing the raw data with a coefficient of determination R²>0.99. The effective oxygen

permeation through polymeric samples can be calculated with this function. A pressure

hysteresis test was undertaken but no hysteresis was found. No temperature dependence

of the oxygen sensor signal was observed in the range of between 20°C and 30°C. This

study presents for the first time data showing the oxygen permeation rates through a low

density polyethylene film under pressures up to 2,000 bars. It could be shown that the

oxygen permeation coefficient is reduced by a factor of between 35 and 70 at 2,000 bars

applied pressure compared to the value at atmospheric pressure at 23°C. These results are

in good accordance with the findings for the permeation of aromatic compounds at similar

pressure conditions. The lowered permeability was attributed to the reduction of free

volume and the decreased chain mobility in polymers. The effect is expected to be more

severe for polymers with initial higher free volume compared to polymers with higher

density.


124

In-situ measurement of oxygen concentration under high

pressure and the application to oxygen permeation through

polymer films

JULIA STERR1, KATHARINA RÖTZER1, KATHRIN WECK1, ANDREAS LEONHARD KARL

WIRTH1, BENEDIKT STEFAN FLECKENSTEIN1 AND HORST-CHRISTIAN LANGOWSKI1,2


2Fraunhofer Institute for Process Engineering and Packaging, Freising, Germany

Published online 21 September 2015 in Journal of Chemical Physics† (2015); Volume

143, pp.114201; DOI: 10.1063/1.4931399

Abstract

Up until now, gas permeation through polymers under high pressure has not been able to

be measured continuously. The combination of a special high pressure cell and a

commercially available fluorescence-based oxygen measurement system allows in-situ

monitoring of oxygen permeation through a polymer sample under pressure in an aqueous

environment. The principle of the oxygen sensor is based on dynamic fluorescence

quenching and measurement of the fluorescence decay time. It was observed that the

decay time increases non-linearly with the applied pressure and hence the displayed

oxygen concentration has to be corrected. This deviation between the measured and the

real concentration depends not only on the pressure but also on the absolute oxygen

concentration in the water. To obtain a calibration curve, tests were performed in the

pressure range between 1-2,000 bars and initial oxygen concentrations in the range

between 40-280 µmol/L. The polynomial calibration curve was of the fourth order,

describing the raw data with a coefficient of determination R²>0.99. The effective oxygen

permeation through polymeric samples can be calculated with this function. A pressure

hysteresis test was undertaken but no hysteresis was found. No temperature dependence

of the oxygen sensor signal was observed in the range between 20°C and 30°C. This study

presents for the first time data showing the oxygen permeation rates through a

polyethylene film in the pressure range between 1 to 2,000 bars at 23°C.

† Permission for the re-use of the article granted by AIP Publishing LLC.


125

Content Publication III

4 In-situ measurement of oxygen concentration under high pressure and the application

to oxygen permeation through polymer films ............................................................... 123

4.1 Introduction .................................................................................................... 126

4.2 Materials and Methods ................................................................................... 129

4.2.1 Hysteresis, reproducibility and temperature dependence ....................... 131

4.2.2 Method for determining the dependence of the oxygen response signal on

the applied pressure ............................................................................................... 131

4.2.3 Oxygen permeation experiments ............................................................ 132

4.4 Results and Discussion .................................................................................. 133

4.4.1 Hysteresis, reproducibility and temperature dependence ....................... 133

4.4.2 Response correction ................................................................................ 134

4.4.3 Oxygen permeation experiments ............................................................ 136

4.5 Conclusions .................................................................................................... 140



126

4.1 Introduction

High pressure processing is an innovative and already well-established technique for

prolonging the shelf life of temperature-sensitive foods such as fresh juices, guacamole

and seafood. Pressures up to 8,000 bars are used. Furthermore, an increasing number of

foods such as ham and cold meat are being packed into packaging having a modified

atmosphere. In order to understand the complex processes that occur in food at high

pressure, it is very important to monitor the chemical, physical, biological and other

processes in-situ. Several attempts have been made to carry out in-situ optical

measurements to monitor changes to food systems, enzymes and microorganisms under

high pressure processing. TORRES, VELAZQUEZ (2005) described a prototype for in-situ

optical polarisation measurements for study processing effects on foods. OGER ET AL.

(2006) developed a diamond anvil cell for observing the growth and fate of microbes

using confocal microscopy and STIPPL ET AL. 2004 described a method for the optical in-

situ measurement of pH. A few other in-situ studies have been performed, but MARTÍNEZ-

MONTEAGUDO, SALDAÑA (2014) mentioned in their review article on “Chemical

Reactions in Food Systems at High Hydrostatic Pressure” that many mechanisms are not

yet fully understood due to the lack of in-situ measurement of chemical reactions. In

particular, the monitoring of in-situ oxygen reactions could help to better understand the

processes that take place under pressure.

Up until now, the permeation of gases through polymer and multilayer films in modified

atmosphere packaging has not been able to be measured. Knowledge of this is very

important because increased amounts of gas dissolved in the polymer can lead to damage

to the polymer packaging after a sharp pressure drop. BOYER ET AL. (2007) described this

phenomenon in their work as “explosive decompression failure” (XDF). The estimation

of gas sorption and diffusion under pressure can help better understand this phenomenon

and make predictions about its occurrence. Some papers have dealt with the sorption of

(supercritical) gases in polymers under pressure using gravimetric methods (LIN ET AL.

2010) or have analysed the amount of gas entering a polymer under controlled

temperature and pressure conditions using a vibrating-wire detector and a pVT technique

(BOYER ET AL. 2009). However, no in-situ oxygen permeation or oxygen reaction

measurements have been performed. In previous studies the permeation of flavour

compounds through polymers was investigated by RICHTER ET AL. 2010. By applying


127

pressure up to 2,000 bars a reduction in the permeation coefficient of ß-ionone through a

low density polyethylene (PE-LD) film was observed by a factor of 55. A reduction by a

factor of six to seven was found in the permeation coefficient of various aroma

compounds (benzoic acid, raspberry ketone and carvacrol) through a polyamide 6 (PA6)

film.

In order to study oxygen permeation through a polymer film, a commercially available

sensor system, a so called optode, based on the principle of dynamic fluorescence

quenching was used for our work. The application of fluorescence lifetime based optodes

in aquatic systems is already used in marine biology and chemistry (TENGBERG ET AL.

2003). Although the measurement of dissolved oxygen is a well-known and often used

technique, the systems are however only designed to operate for low pressure processes.

Inside the sensor spot which is used for dynamic fluorescence measurements, a

fluorophore is embedded in a matrix. This is excited to the electronic singlet state on

illumination with sinusoidally modulated blue light (HUBER 2011; JORGE ET AL. 2004;

MCEVOY ET AL. 1996). Upon fluorescence, the fluorophore returns to the energetic

ground state. In the presence of oxygen, a well-known quencher, collision-induced energy

transfer takes place. The oxygen is excited to its singlet state and the amount of

fluorescence radiation is reduced. This quenching process is described by the Stern-

Volmer Equation 4-1 (STERN, VOLMER 1919):

Equation 4-1

@$@ = A$A = 1 + �CD�EF

The equation indicates that the ratio of the fluorescence light intensity in the absence (I0)

and in the presence (I) of a quencher is proportional to the oxygen partial pressure pO2

and the Stern-Volmer quenching constant KSV. The constant KSV is, according to WELLER

1959 (Equation 4-2), proportional not only to the sum of the molecular radii a and the

number of molecules per millimole N’, but also to the sum D of the diffusion coefficients

of oxygen and fluorophore. As diffusion coefficients are pressure sensitive, this might

explain the observed influence of increasing pressure on the concurrently measured

reduction in fluorescence intensity.

Equation 4-2

�CD = 4HI#JKA$


128

The equipment used in this study determines the luminescence lifetime τ of a fluorophore,

the time for which it is in the excited state. The typical lifetime τ0 of a fluorophore in the

absence of a quencher is in the range of 10 ns to 10 µs (LAKOWICZ 2006; MORRIS ET AL.

2007b). The advantage of measuring the lifetime instead of the intensity is that it is

independent of photobleaching or other external factors such as changes in the light

source intensity or turbidity of the media (HUBER 2011). Additionally, the measuring

method is not oxygen-consuming and shows no sensitivity to stirring (HOLST ET AL. 1995;

TENGBERG ET AL. 2003). Upon excitation of the optode with sinusoidally modulated light

the fluorescence lifetime can be related to a phase delay ϕ in the luminescent emission

relative to the extinction with the modulation frequency fmod (Equation 4-3) (JORGE ET AL.

2004):

Equation 4-3

tan[O] = 2H ��PA

During our experiments it was observed that under identical oxygen concentrations the

fluorescence signal from the sensor spot was strongly dependent on the pressure. Higher

pressure values lead to increasing phase angle ϕ and therefore to decreasing displayed

oxygen concentration cdO2, while the “true” oxygen concentration does not change. The

absence of outgassing phenomena observed through the sapphire window also proves the

constancy of the oxygen concentration. TENGBERG ET AL. (2006) tested a lifetime-based

optode (like the optode used in our studies) at high pressures up to 400 bars. They found

a linear reduction of the sensor response of about 4% per 100 bars. This result could not

be confirmed in our studies, where a non-linear decrease was observed. A reason for the

discrepancy may be, that over a smaller pressure range (3 - 400 bars) the reduction is

approximately linear, but is non-linear over a larger pressure range. Tests were therefore

performed to determine the correlation of the pressure dependent change of the phase

angle ϕ. A test was also undertaken to see whether there was an additional temperature

effect on the signal in the temperature range from 23°C to 29°C. Furthermore a hysteresis

test and reproducibility test were carried out. The results were fitted to a polynomial

function transcribing the raw data to real oxygen concentration values. This study presents

for the first time results for the oxygen permeability of a polyethylene film up to a pressure

of 2,000 bars. Other time-resolved experiments involving oxygen-consuming or oxygen-


129

forming reactions under pressure (such as microbiological respiration or enzymatic

reactions) could be undertaken using this measurement method.

4.2 Materials and Methods

A high pressure vessel from SITEC (Sieber Engineering AG, Maur, Switzerland) was

used for the experiments. This was originally built for studying aroma compound

migration and permeation (RICHTER, LANGOWSKI 2005). Inside the pressure vessel a

polymer sample (ø = 0.7 cm) can be fixed with a retainer (Figure 4-1), so separating the

pressure cell into two chambers. The upper chamber has a volume of 5.4 cm³, the bottom

chamber a volume of 6.7 cm³. Both chambers can be filled and drained separately with

different fluids (e.g. oxygen saturated water and water with reduced oxygen

concentration). The temperature inside the vessel was measured with a thermocouple (Fe-

CuNi Type J) and could be regulated from 20°C to 50°C with the help of a heating jacket

mounted around the vessel. Both chambers were pressurised simultaneously by a

common manual spindle pump. The maximum pressure was 2,500 bars and the pressure

build-up and pressure release rates were around 2,000 bar/min. A pressure sensor made

by EBM Brosa Messgeräte (Tettnang, Germany) was integrated into tubing next to the

vessel. The pressure transmitting fluid was the sample water itself. In the bottom chamber

a sensor spot was fixed on the inside of a sapphire window (thickness 1 cm) for optical

measurement of the fluorescence quenching by the oxygen present in the cell. The sensor

spot PSt3 is commercially available from PreSens - Precision Sensing GmbH

(Regensburg, Germany). It operates within a specific measurement range (0 to 1,400

µmol/L dissolved oxygen) and has a resolution of 6%. The sensor spot is illuminated from

the outside with blue light by an optical fibre connected to the Fibox 3 LCD Trace

measurement system (PreSens - Precision Sensing GmbH Regensburg, Germany).

The oxygen measurement system was calibrated before starting the experiments and after

every 100,000 data points. A two-point calibration was carried out at atmospheric

pressure by first adding sodium sulphite (Carl Roth GmbH + Co. KG, Karlsruhe,

Germany) to the water to create an oxygen-free environment. The second calibration point

was taken with 100% air-saturated water. At 23°C and 1,013 mbar pressure, air-saturated

water contains 280 µmol oxygen per litre water (calculated using Equation 4-7, SANDER

2015).


130

Figure 4-1 Schematic presentation of the pressure cell [according to RICHTER, LANGOWSKI (2005); Copyright Tobias Richter. Reproduced with permission of Der Weihenstephaner 44, 85 (2005).]

For the experiments, water with reduced oxygen concentration was prepared by boiling

and subsequent cooling water to 23°C in hermetically sealed glass containers. This

enabled oxygen concentrations of down to 40 µmol/L to be reached (approx. 20% of the

oxygen saturation of water under atmospheric conditions). To produce water with

different concentrations of dissolved oxygen, the cooled water was stirred under

atmospheric conditions for varying periods of time. For the preparation of air-saturated

water, the water was stirred for at least 30 minutes. The oxygen concentrations were

validated before every experiment by take measurements with the sensor in the pressure

vessel. All the experiments were performed with demineralised and boiled water to

prevent microbiological contamination. Additionally, the system was cleaned between

every experiment with RIMASAN®-AQ N-34524 (Tensid Chemie GmbH,

Muggensturm, Germany) to avoid microbiological growth and hence oxygen attrition.


131

4.2.1 Hysteresis, reproducibility and temperature dependence

Preliminary experiments were carried out to check the reproducibility and reliability of

the sensor reaction. As the vessel has two sapphire windows it was possible to fix two

sensor spots inside the vessel, one spot on each window. With two independently working

measuring devices the reduction of the two phase angles with the applied pressure could

be monitored independently. Data for testing for hysteresis and temperature dependence

were collected at the same time as the determination of the oxygen response signal

correction.

4.2.2 Method for determining the dependence of the oxygen response signal on the

applied pressure

For these experiments, water with varying initial oxygen concentrations (40 µmol/L <

ciO2 < 260 µmol/L) was prepared and filled into the vessel. To check whether air bubbles

remained in the system, the pressure was first increased to 20 bars and held for a few

minutes (Figure 4-2 and Figure 4-3). At this low pressure the sensor works within

standard parameters and additional dissolved oxygen immediately increases the displayed

oxygen concentration. In such an eventuality, the experiment was interrupted and

restarted. Otherwise the pressure was increased gradually up to a maximum pressure of

2,000 bars in steps of 200 bars (Figure 4-2). After every step the pressure was held for a

few minutes until the response signal was stable. The pressure release was also performed

in steps to check for possible pressure hysteresis. The temperature was kept constant in

each experiment at respectively 23°C, 25°C, 27°C and 29°C. In order to obtain a

correlation curve describing the raw data as real values, a mean oxygen concentration at

every pressure step as a function of the initial oxygen concentration ciO2 was calculated.

Curve fitting was performed with the Curve Fitting Toolbox of MATLAB R2013b. A

polynomial function was chosen as the regression model and the order of the polynomial

function was selected, considering the sum of squared errors due to error (SSE) and the

coefficient of determination.


132

Figure 4-2 Stepwise increase in pressure leads to a simultaneous decrease in the displayed oxygen concentration cdO2.

4.2.3 Oxygen permeation experiments

For the permeation measurements, the bottom chamber was filled with water having a

reduced concentration of oxygen. After mounting the polymer sample, the upper chamber

was filled with air-saturated water. To overcome the low diffusion coefficient of oxygen

in water, two magnetic stirrers (one in each chamber) ensured the homogeneous

distribution of oxygen in the water during the permeation experiments. All the permeation

experiments were carried out with low density polyethylene (PE-LD) film samples of 50

µm and 100 µm thickness made from additive free granulate (Lupolen 3020 K by

LyondellBasell, Wesseling, Germany) on a Collin lab extrusion line at the Fraunhofer

Institute for Process Engineering and Packaging IVV in Freising, Germany.


133

4.4 Results and Discussion

4.4.1 Hysteresis, reproducibility and temperature dependence

The results in Figure 4-3 demonstrate that the sensor response to the oxygen concentration

is reproducible, namely the two independently operating sensors show identical signals

at the same time within the working range of the sensor (the maximum deviation of the

measured oxygen concentration was about 6%).

Figure 4-3 Comparison of two independently operating sensors. Nominal oxygen concentration and applied pressure, at an absolute O2 concentration of 237 µmol/L.

Neither a hysteresis effect nor a temperature dependence could be detected. Figure 4-4

shows the oxygen concentration as a function of the pressure p for three representative

trials with varying absolute oxygen concentrations caO2 (pink, blue and black). The

different symbols represent pressure build-up and pressure release. It can be clearly seen

that the oxygen signal decreases non-linearly with increasing pressure and vice versa and

that the values follow the same curve for pressure build-up and release. Additionally, it

should be mentioned that the initial and final values at atmospheric pressure lie in the

same range. Thus, it can be concluded that there is no hysteresis effect.


134

Figure 4-4 Hysteresis test. Nominal oxygen concentration as a function of pressure at different absolute oxygen concentrations caO2.

To examine the temperature dependency, the relative deviation Δcd(T) of the data from

the experiments at 23°C and 29°C were plotted as a function of the pressure p and absolute

oxygen concentrations caO2 (Figure 4-5). A maximum deviation of 6% was found for the

lowest absolute oxygen concentration (50 µmol/L) at atmospheric pressure and this is

comparable with the measurement error of the sensor. Based on these findings, there was

assumed to be no additional temperature dependence in the 20 to 30°C range at high

pressures. A temperature influence was therefore not taken into account for further

measurements. No predictions can be made, however, for temperatures above or below

this range. The maximum adiabatic heating was found to be less than 1°C at the maximum

pressure build-up rate (1,000 bar/min) and therefore can also be neglected.

4.4.2 Response correction

The correlation of the displayed and real oxygen concentration as a function of the

pressure p and absolute oxygen concentration caO2 was described by a polynomial

function of the fourth order (Equation 4-4) with a coefficient of determination R²>0.99

within the data range.


135

Equation 4-4

(�, <QF) = <RQF= I00 + I10 ∗ � + I01 ∗ <QF + I20 ∗ �F + I11 ∗ � ∗ <QF + I02∗ <QFF + I30 ∗ �U + I21 ∗ �F ∗ <QF + I12 ∗ � ∗ <QFF + I03 ∗ <QFU+ I40 ∗ �V + I31 ∗ �U ∗ <QF + I22 ∗ �F ∗ <QFF + I13 ∗ � ∗ <QFU+ I04 ∗ <QFV

The terms pn and cnO2 in the function stand for the normalised raw data values for the

pressure p and displayed oxygen concentration cdO2 (Equation 4-5 and Equation 4-6). For

fitting with a high-order polynomial, the data should be normalised by centring it at zero

mean and scaling it to unit standard deviation.

The values of the coefficients used for Equation 4-4 can be found in Table 4-1.

Figure 4-5 Test for temperature dependence. Relative deviation of data at 23°C and 29°C as a function of pressure p and absolute oxygen concentration caO2.


136

Equation 4-5

� = (� − 1000 ℎ�I)638.3 ℎ�I

Equation 4-6

<QF = (<PQF − 110.3 μ]^_/a)55.47μ]^_/a

Table 4-1 Coefficients for Equation 4-4 with 95% confidence interval.

Coefficient Value 95% confidence

interval 95% confidence

interval a00 152.3 151.5 153.1 a10 29.03 27.98 30.09 a01 76.61 75.3 77.92 a20 2.679 1.299 4.129 a11 19.42 17.64 21.2 a02 2.162 0.6126 3.711 a30 1.1 0.5822 1.617 a21 5.932 4.827 7.036 a12 8.968 7.441 15.5 a03 5.219 4.226 6.212 a40 0.03086 -0.5146 0.5763 a31 0.9104 0.2155 1.605 a22 3.385 2.056 4.717 a13 4.804 3.11 6.498 a04 1.799 0.9609 2.637

An example of the verification of the function is shown in Figure 4-6, whereby the data

of Figure 4-4 were recalculated with Equation 4-4, Equation 4-5 and Equation 4-6. The

fact that the values do not change with pressure demonstrates the accuracy of the function.

4.4.3 Oxygen permeation experiments

For calculation of the permeability values, the initial slope of the oxygen concentration at

the beginning of the permeation process was used, representing the permeation rate at

maximum concentration difference. The polymer was already saturated with oxygen at

atmospheric pressure and a possible time lag was too short to be observed. Figure 4-7

shows, for example purposes, the oxygen concentration at the bottom chamber of the

pressure vessel for permeation measurement through a polyethylene film (50 µm) at 500


137

bars and 23°C. The gradient of the oxygen concentration between the two chambers was

212 µmol/L. The increase in the oxygen concentration in the first three hours

approximately represents the steady state of permeation.

Figure 4-6 Recalculated raw data giving real values for the hysteresis test shown in Figure 4-4.

Figure 4-7 Permeation measurement through a polyethylene film of 50 µm thickness at 23°C and 500 bars pressure. The oxygen concentration was measured at the bottom chamber of the pressure vessel.


138

The oxygen concentration values were converted to absolute amounts of oxygen using

the volume of the bottom chamber (6.7 cm³) and the molar volume of oxygen (22.4

L/mol). For gas permeation measurements, it is traditional to use a volume unit (cm³) at

standard temperature and pressure STP (273K and 1,013.25 mbar). The oxygen

concentration was converted into a partial pressure of oxygen (mbar) using Henry’s law

(< = d ∗ �). Values of Henry’s constant kH,cp can be found in SANDER (2015), giving a

value for oxygen in pure water of:

Equation 4-7

de,R� = 1.3 ∗ 10fU ]^_ ��a ∗ 1013.25 ]gI; ∗ %�� h1700 � ∗ �1i − 1

298��k

The measured values for the oxygen permeation rate through polyethylene films are then

displayed in cm³ [STP] /(m²*d). With further normalisation to a film thickness of 100 µm

and to a partial pressure oxygen gradient of 1 bar, namely the driving force for the

permeation, the permeation coefficient P is obtained in cm³ [STP] 100 µm / (m²*d*bar).

The normalised results for two polyethylene films of different thickness (50 and 100 µm)

are shown in Figure 4-8 and Table 4-2. The error bars indicate the 95% confidence

interval. The fact that the decrease in the permeation coefficient on pressure increase is

very similar for both films demonstrates clearly that the pressure effect is independent of

the film thickness. Apparently, there is an exponential correlation between the oxygen

permeation coefficient and the applied pressure. Comparison of the oxygen permeation

coefficient at atmospheric pressure with data from literature shows good agreement.

Typical permeation coefficients for oxygen in low density polyethylene in a gaseous

environment range from 1,800 to 2,000 (cm³(STP) 100 µm) / (m² d bar) (LANGOWSKI

2008).


139

Figure 4-8 Oxygen permeation coefficient of polyethylene films (50 and 100 µm thickness) at 23°C for pressures up to 2,000 bars

Table 4-2 Pressure dependence of the oxygen permeation coefficient in PE-LD at 23°C

Film thickness [µm]

Pressure [bar]

Oxygen permeation coefficient

l <]U[,i�]]F ∗ m ∗ gI; [100μ]]n

95% confidence interval

100 µm 1 1,540 165 500 - - 1,000 160 44 1,500 36 6 2,000 44 13 50 µm 1 2,090 75 500 307 146 1,000 126 24 1,500 83 20 2,000 29 17


140

In relation to the values at atmospheric pressure, the permeation coefficients of the

polymer films are reduced by factors between 35 and 70 at pressures up to 2,000 bars.

These findings are in very good agreement with the results of RICHTER ET AL. (2010) and

RICHTER (2011), where a lowering of the permeation coefficient of β-ionone through a

low density polyethylene film by a factor of 55 was observed. These results confirm that

the lower permeability at high pressure can be attributed to the reduction in the free

volume and the decreased chain mobility in polymers. This is in agreement with the

findings for the permeation of model substances (ß-ionone, benzoic acid, carvacrol and

raspberry ketone) through polyamide 6 which indicated a lowering of the permeation

coefficient by only a factor of 7 (RICHTER 2011; RICHTER ET AL. 2010). Apparently, an

increase in pressure has more influence on the reduction of the permeation for polymers

having a high free volume and high chain mobility, such as low density polyethylene.

Finally, it should be mentioned that the process conditions easily exceed the critical point

of oxygen (154K, 50 bars). Therefore, further permeation measurements using this new

method would be worthwhile in the pressure range around the critical point of oxygen

and using different types of polymers.

4.5 Conclusions

The principle of fluorescence quenching was used for in-situ measurement of the

concentration of oxygen dissolved in water under high pressure and was applied for

permeation measurements. At constant real oxygen concentration, a non-linear

correlation was observed between the fluorescence quenching, and hence the displayed

oxygen concentration, and the applied pressure. The correlation can be formally described

by a polynomial function of the fourth order. This function enables correction of the

displayed oxygen concentrations and the calculation of oxygen permeation rates under

high pressure conditions. The measurement method is reproducible and the data show no

hysteresis effect and no temperature dependence in the range of 23°C to 29°C. Using this

experimental setup it is possible to measure oxygen transport processes (such as

permeation through polymers) at high pressure. It was shown that the oxygen permeation

coefficient of a low density polyethylene film is reduced by a factor of between 35 and

70 at 2,000 bars applied pressure compared to the value at atmospheric pressure.


141

In general, the experimental system has potential for addressing a host of other scientific

problems and issues, for example measurement of the oxygen uptake of microorganisms

under pressure and monitoring oxygen in reactions involving antioxidants, free radicals

and isomerization processes.


142





10.1002/app.25085.

BOYER, S. A. E.; YAMADA, T.; YOSHIDA, H.; GROLIER, J.-P. E. (2009): Modification of

molecular organization of polymers by gas sorption: Thermodynamic aspects and

industrial applications. In Pure and applied chemistry 81 (9), pp. 1603–1624. DOI:

10.1351/PAC-CON-08-11-09.

HOLST, G. A.; KUEHL, M.; KLIMANT, I. (1995): Novel measuring system for oxygen

micro-optodes based on a phase modulation technique. In Annamaria V. Scheggi

(Ed.): Chemical, Biochemical, and Environmental Fiber Sensors VII, vol. 2508.

European Symposium on Optics for Environmental and Public Safety: International

Society for Optics and Photonics, pp. 387–398.

HUBER, C. (2011): Appendix. Basics in optical sensing of oxygen. Edited by PreSens

Precision Sensing GmbH.

JORGE, P. A. S.; CALDAS, P.; ROSA, C. C.; OLIVA, A. G.; SANTOS, J. L. (2004): Optical

fiber probes for fluorescence based oxygen sensing. In Sensors and Actuators B:

Chemical 103 (1), pp. 290–299.

LAKOWICZ, J. R. (2006): Principles of fluorescence spectroscopy. 3 volumes. New

York: Springer.

LANGOWSKI, H.-C. (2008): Permeation of gases and condensable substances through

monolayer and multilayer structures. In O. G. Piringer, A. L. Baner (Eds.): Plastic

packaging. Interactions with food and pharmaceuticals. 2nd ed. Weinheim,

Chichester: Wiley-VCH, pp. 297–348.

LIN, S.; YANG, J.; YAN, J.; ZHAO, Y.; YANG, B. (2010): Sorption and diffusion of

supercritical carbon dioxide in a biodegradable polymer. In J. of Macromolecular

Sc., Part B 49 (2), pp. 286–300. DOI: 10.1080/01495930903352308.


143

MARTÍNEZ-MONTEAGUDO, S. I.; SALDAÑA, M. D. A. (2014): Chemical reactions in food

systems at high hydrostatic pressure. In Food Engineering Reviews 6 (4), pp. 105–

127.

MCEVOY, A. K.; MCDONAGH, C. M.; MACCRAITH, B. D. (1996): Dissolved oxygen

sensor based on fluorescence quenching of oxygen-sensitive ruthenium complexes

immobilized in sol–gel-derived porous silica coatings. In Analyst 121 (6), pp. 785–

788.

MORRIS, K. J.; ROACH, M. S.; XU, W.; DEMAS, J. N.; DEGRAFF, B. A. (2007):

Luminescence lifetime standards for the nanosecond to microsecond range and

oxygen quenching of ruthenium (II) complexes. In Analytical Chemistry 79 (24),

pp. 9310–9314.

OGER, P. M.; DANIEL, I.; PICARD, A. (2006): Development of a low-pressure diamond

anvil cell and analytical tools to monitor microbial activities in situ under controlled

P and T. In Proteins Under High Pressure 1764 (3), pp. 434–442. DOI:

10.1016/j.bbapap.2005.11.009.



RICHTER, T.; LANGOWSKI, H.-C. (2005): Hochdruckbehandlung von verpackten

Lebensmitteln - Beurteilung flexibler, polymerer Verpackungsmaterialien für den

Einsatz in der Hochdruckpasteurisation. In Der Weihenstephaner 44 (2), pp. 83–86.



566. DOI: 10.1080/08957959.2010.531722.

SANDER, R. (2015): Compilation of Henry’s law constants (version 4.0) for water as

solvent. In Atmos. Chem. Phys. 15 (8), pp. 4399–4981. DOI: 10.5194/acp-15-4399-

2015.

STERN, O.; VOLMER, M. (1919): On the quenching-time of fluorescence. In Phys. Z. 20,

pp. 183–188.

STIPPL, V. M.; DELGADO, A.; BECKER, T. M. (2004): Development of a method for the

optical in-situ determination of pH value during high-pressure treatment of fluid


144

food. In Innovative Food Science & Emerging Technologies 5 (3), pp. 285–292.

DOI: 10.1016/j.ifset.2004.05.002.

TENGBERG, A.; HOVDENES, J.; ANDERSSON, H.; BROCANDEL, O.; DIAZ, R.; HEBERT, D.

et al. (2006): Evaluation of a lifetime-based optode to measure oxygen in aquatic

systems. In Limnol. Oceangr. 4, pp. 7–17.

TENGBERG, A.; HOVDENES, J.; BARRANGER, D.; BROCANDEL, O.; DIAZ, R.; SARKKULA,

J. et al. (2003): Optodes to measure oxygen in the aquatic environment. Absolute

oxygen concentrations with high precision without repetitive calibrations; long-term

stability. In Sea Technology 44 (2), pp. 10–16.

TORRES, J. A.; VELAZQUEZ, G. (2005): Commercial opportunities and research

challenges in the high pressure processing of foods. IV Iberoamerican Congress of

Food Engineering (CIBIA IV). In Journal of Food Engineering 67 (1-2), pp. 95–112.

WELLER, A. (1959): Outer and inner mechanism of reactions of excited molecules. In

Discuss. Faraday Soc. 27, pp. 28–33.

PUBLICATION IV CHAPTER 5

145

5 The effect of high pressure processing on tray packages with

modified atmosphere

From previous studies (Chapter 3) it was getting clear, that relevant scientific work is

missing concerning the effect of HPP on tray packages with modified atmosphere. Up

until now scientific information about the influence of gases at high pressure conditions

on the morphology of polymers, the barrier properties and the overall integrity of

polymeric packaging was lacking.

Hence for this study the changes in the morphology of different polymeric materials after

high pressure treatment was determined. The density of polyethylene terephthalate as well

as the crystalline and amorphous content of polyethylene were analysed quantitatively

with Raman spectroscopy. This spectroscopic method allows non-destructive

determinations of the samples with very good spatial resolution.

For the investigations headspace volume (15 to 50% of the total package), headspace gas

composition (air, CO2, N2 and a mixture of 20% CO2/ 80% N2), multilayer structures,

film thickness and flexibility of the lid were varied. Additionally the influence of

headspace gases in HPP on the barrier properties of an organic and an inorganic material

and on the specific as well as overall migration of additives into liquid food simulants has

been studied.

The results show small but technically negligible changes in polymeric structure. An

increased volume of headspace gas seemed to have an influence on the amorphous phase

of polyethylene. The density of polyethylene terephthalate (PET) decreases only slightly

after high pressure treatment. Considering the barrier properties of organic and inorganic

barrier layers it was getting clear that ethylene-vinyl alcohol copolymer (EVOH) is

suitable for the HPP process, whereas metalized layers showed bad performance due to

insufficient mechanical properties. No additional risk by the supercritical CO2 extraction

of packaging components could be detected in the migration studies.

With respect to these outcomes it can be concluded that flexible polymeric packaging is

suitable for the use as modified atmosphere packaging at high pressure treatment.


146

The effect of high pressure processing on tray packages with

modified atmosphere

JULIA STERR1, BENEDIKT STEFAN FLECKENSTEIN1, HORST-CHRISTIAN LANGOWSKI1,2



Published online 03 June 2014 in Springer Science+Business Media New York 20143:

Food Engineering Reviews (2015); Volume 7, Issue 2, pp. 209-221; DOI: 10.1007/s12393-014-9081-z

ABSTRACT

High pressure processing (HPP) is a small, but growing and profitable technique to extend

the shelf life of high quality food products. As tray packages with modified atmosphere

(MAP) provide better product presentation and also improved product protection, it seems

logical to combine HPP with MAP. Due to the fact that the used gases like oxygen,

nitrogen and carbon dioxide are in a supercritical state during high pressure processing,

it is important to give extra attention to these conditions. Several studies were performed

dealing with the influence of high pressure on the mechanical and/or barrier properties of

packages and polymers. However, little information is known about the combined effect

of MAP gases and high pressure on packaging. This study hence evaluates the influence

of headspace gases on the morphology and integrity of packages subjected to HPP.

Pressure related changes of the morphology of polymers as commonly used for packaging

will be shown. Density as well as crystalline and amorphous content were analyzed

quantitatively with Raman spectroscopy by varying headspace volume (15 to 50% of the

total package), headspace gas composition (air, CO2, N2 and a mixture of 20% CO2/ 80%

N2), multilayer structures, film thickness and flexibility of the lid. Additionally the

influence of headspace gases in HPP on the barrier properties of an organic and an

inorganic material and on the migration of additives into liquid food simulants has been

studied.

3 Permission for the re-use of the article is granted by Springer


147

Content Publication IV

5 The effect of high pressure processing on tray packages with modified

atmosphere ..................................................................................................................... 145

5.1 Abbreviations ................................................................................................. 148

5.2 Introduction .................................................................................................... 149

5.3 Materials and Methods ................................................................................... 150

5.3.1 Monolayer materials ............................................................................... 150

5.3.2 Multilayer films in model packages ....................................................... 151

5.3.3 Layer thickness ....................................................................................... 152

5.3.4 Raman spectroscopic measurements ...................................................... 153

5.3.5 Oxygen permeation measurement .......................................................... 156

5.3.6 Migration studies .................................................................................... 156

5.4 Results and Discussion .................................................................................. 158

5.4.1 Crystallinity and density changes of monolayer materials ..................... 158

5.4.2 Multilayer films in model packages and real packages .......................... 161

5.4.2.1 Influence of layer thickness............................................................. 161

5.4.2.2 Influence of headspace volume and gas composition ..................... 163

5.4.3 Oxygen permeation of multilayer lid films ............................................ 164

5.4.3.1 EVOH barrier layer ......................................................................... 164

5.4.3.2 Inorganic barrier layer ..................................................................... 164

5.4.4 Migration ................................................................................................ 165

5.5 Conclusions .................................................................................................... 167

5.6 Acknowledgments ......................................................................................... 168



148

5.1 Abbreviations

Adh. Adhesive

Al Aluminum

CCD Charged coupled device

DSC Differential scanning calorimetry

EVA Ethylene-vinyl acetate copolymer

EVOH Ethylene vinyl alcohol copolymer

FT-IR Fourier transform infrared spectroscopy

HPP High pressure processing

MAP Modified atmosphere packaging

PA Polyamide

PA6 Polyamide 6

PE-HD High density polyethylene

PE-LD Low density polyethylene

PE-LLD Linear low density polyethylene

PE-MD Medium density polyethylene

PET Poly(ethylene terephthalate)

o PET Poly(ethylene terephthalate), biaxially oriented

a PET Amorphous Poly(ethylene terephthalate)

PP Polypropylene

PTFE Polytetrafluoroethylene

SiOx Silicon oxide

XRD X-ray diffraction


149

5.2 Introduction

High pressure processing (HPP) of food serves the rising demands of the consumers for

natural, low processed food with less additives. Compared to a thermal treatment fresh

and safe products with prolonged shelf life can be created and allow to fulfil the

consumers requests. In HP batch processes packed products are treated to avoid

recontamination or contact with the pressure transmitting fluid. But the packaging has to

withstand the pressurizing process and the producer has to ensure that the material still

maintains its original properties and fulfils legal requirements. Several investigations

have been made about changes generated by high pressure inside polymers and packages.

Both FLECKENSTEIN ET AL. (2014) and JULIANO ET AL. (2010) give an overview about

publications related to this topic. Most of the cited studies, however, just deal with films

and/or pouches used for vacuum packages without headspace. However in industrial

applications many products (e.g. meat products like ham or sausages) are packed in tray

/ lid combinations with a headspace of modified atmosphere. Nitrogen (N2) and carbon

dioxide (CO2) are the gases most often used in MAP. But to our knowledge no paper

tackled the effect of headspace gases on the morphology, permeability and migration

properties of polymers used in the food industry. Additionally it has to be taken into

account that common headspace gases undergo a transition to the supercritical state in

HPP (critical points: oxygen 154 K/51 bar; nitrogen 126 K/33,8 bar; CO2

303,2 K/73,81 bar) and, therefore, show changed properties like a better solubility in

other media (LAX ET AL. 1998). Although the critical temperature of carbon dioxide is as

high as 30°C, pressure build up in almost adiabatic conditions will always raise the

temperature above the critical point of CO2. Supercritical carbon dioxide is well known

to be an efficient extraction solvent (e.g. for decaffeination). An increased extraction of

polymer additives in high pressure treatment is a possible risk that should be considered.

Also little information is available about the influence of supercritical fluids on the

structure of polymers.

Another consequence of headspace gases in combined HPP and MAP are bubbles, blisters

and so called White Spots inside the films as observed after HP treatment. BOYER ET AL.

(2007) identified the phenomenon of “explosive decompression failure” in plant

components of the petroleum industry and explained it by the immediate and rapid

expansion of gas dissolved in polymeric parts upon pressure release. This unwanted effect


150

may also delaminate films or destroy barrier layers (BULL ET AL. 2010; FAIRCLOUGH,

CONTI 2009; RICHTER ET AL. 2010). So the aim of this study is to analyse the influence of

headspace gases in packages for HPP on the morphology of polymers and the integrity of

packages.

First the impact of HPP on the morphology on the polymeric material was investigated,

because the structure strongly affects the properties of the films. Structural modifications

would have a direct consequence on the polymer characteristics, like permeability,

appearance or mechanical properties. Morphological analyses were performed using

Raman spectroscopy with a high spatial resolution in a confocal laser scanning

microscope, further called Raman microscopy. At a lateral and vertical spatial resolution

down to 260 nm, physical information of the inside of a transparent material can be

obtained non-destructively in three dimensions, e.g. for the identification of inclusions or

for the identification and determination of the individual thickness of layers in a multi-

layer system (HOLLRICHER, IBACH 2011).

To separate the influence of high pressure process parameters (pressure, time and

temperature) from packaging design parameters (headspace volume, gas composition,

thickness and layer sequence of films) on the polymer structure, the results obtained for

the single-layer samples had to be compared with those from examinations with tray

packages filled with a model food product under modified atmosphere. In addition,

possible changes in the oxygen permeability and of the specific and overall migration

were investigated.

5.3 Materials and Methods

5.3.1 Monolayer materials

For the analysis of pressure dependent changes in crystallinity and density with Raman

microscopy, different single layered films were used as basic reference samples. The

samples were treated at different hydrostatic pressures, temperatures and periods of

exposure. Sheets of approximately 20x30 cm² were cut from samples PE-LD2, PET1 and

PET2 and pressurized in a NC Hyperbaric apparatus. The samples PE-LD1 and PE-HD

were treated 1-5 days in a high pressure measurement cell from Sitec (RICHTER,

LANGOWSKI 2005). The experiments were part of another scientific study to measure the

in-situ permeation of aroma compounds through polymers under pressure. The


151

experimental setup is described in RICHTER ET AL. (2010) and also in the dissertation of

RICHTER (2011). The duration of these experiments (1-5 days) depended on the

permeation rate of the compounds through the polymer and of the applied pressure. The

influence of the parameters (processing time and pressure) to the values of crystallinity

were compared separately but no influence of time was found. Therefore all results with

different time were pooled together. Due to the small inner volume of the cell, a negligible

adiabatic heating could be achieved. During the pressurizing process all samples were in

direct contact with the pressure transmitting fluid (PE-LD1 and PE-HD: solution of

ethanol at a volume concentration of 5,2% in water; PE-LD2, PET1, PET2: water). For

detailed information about the films and the applied conditions see Table 5-1.

5.3.2 Multilayer films in model packages

For the evaluation of the influence of headspace gases model packages were built. Basis

was a tray from a multilayer film with the layer sequence PET/EVA/PE-MD/EVOH/PE-

MD/PE-LLD (See Figure 5-1 and Table 5-1). The thickness of PET was varied (400 or

600 µm) to check the influence of the total thickness on the overall integrity of the

packaging. Trays were made by vacuum thermoforming in a circular mold of 16 cm

diameter and 3 cm depth. Two lids with different stability to deformation were used:

• An elastic multilayer lid (thickness 120 µm) consisting of PE-

mLLD/Adh./EVOH/Adh./PE-mLLD (Table 5-1).

• A laminated lid made of oPET, vacuum coated with a thin SiOx layer, laminated

to PE-LD, with a low elasticity, i.e. a stretching limit of about 5% in elongation

until loss of barrier properties (Table 5-1).

The trays were filled with an agar gel (Agar-Agar Kobe I, 15g/L, Carl Roth, Karlsruhe,

Germany) to simulate a water based, compact food product. With the agar gel it was

possible to vary the headspace volume reproducibly over a wide range (volume fraction

of the headspace: 15, 30 and 50%). The atmosphere inside was either air, 100% CO2,

100% N2 or a mixture of 20% CO2 with 80% N2. Every combination (tray, lid, headspace

volume and gas composition) was prepared in triplicate. The high pressure treatment was

performed at 600 MPa for 5 min at room temperature (Table 5-1).

For the investigations with MAP, common industrial packages were also used, supplied

by a company from the project consortium. The tray material consists of PET/PEcoex


152

(210 µm/34 µm). It was previously investigated in the study of RICHTER, LANGOWSKI

(2005) with respect to its performance in HPP. Examinations of a microtome cut with a

polarizing microscope and the compared investigations with Raman spectroscopy showed

that the PE layer consists of three different types of PE (see also Figure 5-2). The trays

(dimension of 21 cm x 18 cm x 0,5 cm) were filled with 100 g of sliced dried prosciutto

at a headspace gas composition of 20% CO2 an 80% N2 and HP treated under industrial

conditions (600 MPa, 2 min, 5°C). After HPP, all packages were stored at ambient

conditions, i.e. room temperature. Detailed information to find in Table 5-1.

5.3.3 Layer thickness

To find out whether differences in the thickness of the material have a measurable impact

on the polymer structure and therefore favor the formation of XDFs, the morphology of

polyethylene in multilayer trays was investigated via Raman spectroscopy. The inner side

of the model tray contains two (Figure 5-1 left side) and of the industrial tray three (Figure

5-2) different polyethylene types. With confocal Raman spectroscopy the thickness of the

different PE layers could be estimated based on a depth scan of the film. The estimation

of layer thickness via confocal Raman microscopy has a limited accuracy. Although an

oil immersion objective was used in this study, the results should be considered more as

a qualitative indication than as a precise measurement of the individual layer thickness.

Figure 5-1. Model tray. Left: Raman depth scan of a model tray (PET/EVA/PE-MD/Adh./EVOH/Adh./PE-MD/PE-LLD) and differences in layer thickness due thermoforming (right side)


153

Figure 5-2. Industrial tray. Optical micrographs of two microtome cuts through white spots in the PE-LD layer in a PET/PEcoex laminate. (Sterr 11/2/2012)

5.3.4 Raman spectroscopic measurements

Raman spectra were recorded using a confocal laser scanning microscope with a green

laser (532 nm) (WITEC Confocal Raman and Atomic Force Microscope Alpha 500). For

the measurements a Nikon oil immersion objective (100x/1.25 Oil∞/0,17 WD 0.18) was

used with oil modified for microscopy (following ISO 8036) from Merck (Darmstadt,

Germany). The objective provides a calculated spatial resolution of less than 0,3 µm. The

scattered light is collected by a spectrograph equipped with a CCD camera and allows a

resolution of the spectral shift of about 2,5 cm-1 in the range from 0 to 3700 cm-1. The

pinhole diameter to the spectrograph is 50 µm.

Due to the confocal setup of the microscope the samples did not have to be cut or prepared

for analysis. For the quantitative determination of crystallinity or density area scans were

taken at the same depth (6 µm underneath the surface). For the analyses in coextruded

polyethylene layers the scanning position of the second and third layer depended on the

overall thickness. It should be noted that changes in the morphology of the polymers were

only compared for samples of identical vertical scan position. Scans were always

performed with a lateral size of 6 µm to 10 µm edge length. Per scan minimum 90 single

spectra/ points per line and 90 lines per image were created. This results in a minimum

number of 8100 spectra for each scan. The average of these spectra was used for further

calculations. Peak fitting was performed with the mathematical program OriginPro8. For

the deconvolution of the spectra of polyethylene the fingerprint region was divided into

three parts. The first part contains four peaks and represents the C-C-stretching vibration.

For this part, approximately from 950 – 1240 cm-1, a Lorentzian function was used to

model the peak shapes. The bands in the second spectral range (1200 – 1400 cm-1),

belonging to the CH2-twisting vibration and the third range from 1390 cm-1 to 1500 cm-1


154

(CH2-bending vibration) were fitted with a Gaussian function (see Figure 5-3) (MAXFIELD

ET AL. 1978). All fits had a minimum coefficient of determination of R²>0,99.

Figure 5-3 Peakfitting of three parts of the fingerprint region of PE. Squares represent raw data. Solid line show fitted data. Dotted lines show fitted bands. (R²>0,99)

Table 5-1 Films, packages, HP-systems and conditions

High pressure systems

Nr. Short name Supplier Pressure build up Maximum adiabatic heating

1 Sitec SITEC-Sieber Engineering AG, Maur, Switzerland

300 MPa/min negligible

2 NC Hyperbaric

NC Hyperbaric, Burgos, Spain

120 MPa/min ~30°C

3 System from project consortium

n/a n/a

High pressure conditions

HP system

HP condition Pressure applied Initial temperature Treatment time

1 A 50 - 200 MPa 23°C 1-5 days

B 50 - 200 MPa 40°C 1-5 days

2 C 300 MPa 23°C 5 min

D 600 MPa 23°C 5 min

E 600 MPa 23°C 60 min

3 F 600 MPa 5°C 2 min

Monolayer films (tested without headspace)

Polymer Thickness Type Supplier HP condition

PE-HD 45 µm - defa-Folien, Lohmar, Germany

A

PE-LD 1 30 µm - defa-Folien, Lohmar, Germany

A/B


155

PE-LD 2 50 µm Lupolen® 3020k LyndellBasell, Ludwigshafen, Germany

D/E

PET 1 23 µm Melinex® 377 Pütz, Taunusstein, Germany

C/D

APET 2 75 µm Hostaphan® RNK Mitsubishi, Wiesbaden, Germany

D/E

Multilayer (packages with headspace)

Package component

Multilayer film spec. Supplier HP condition

Model tray

PET (400 or 600µm)/ EVA (10µm)/ PE-MD (10µm)/Adh./EVOH (7µm)/ Adh./ PE-MD (10µm)/ PE-LLD (13µm)

Mondi Technologies (Gronau, Germany)

D

Elastic lid PE-mLLD (48µm)/ Adh. (9µm)/EVOH (7µm)/Adh. /PE-mLLD (48µm)

Inelastic lid oPET (12 µm) SiOx/Adh./PE-LD(80 µm)

Pouch PE-LD (40µm)/Adh./PA (5µm)/EVOH (2,7µm)/PA (5µm)/Adh./PE-LD (40µm)

Industrial tray PET (210 µm)/PEcoex (34µm) Packer from project consortium

F

Based on relevant literature (see specific references below) on Raman spectroscopy of

polyethylene, the crystalline and amorphous content was determined. The fingerprint

region of the spectrum contains ten peaks, each allocated to different vibrations in the

polymer. A good overview about the assignment of the Raman bands is given by PIGEON

(1991). For the calculation of the crystallinity in polyethylene STROBL, HAGEDORN (1978)

found a proportional correlation between the integrated intensity @oVop R�qr of the band

at 1416 cm-1 (CH2 bending), due to factor group splitting in an orthorhombic crystal

(Equation 5-1). The isotropic amorphous phase is proportional to the intensity @o$s$ R�qr

at 1080 cm-1, according to the C-C stretching vibration (Equation 5-2). An internal

standard ISt to which the intensity of the peaks could be referred to, was found at around

1300 cm-1 and belongs to the C-H twisting vibration. Polyethylene contains not only

crystalline and amorphous phases, but also a “disordered phase of anisotropic nature,

where chains are stretched but have lost their lateral order” (STROBL, HAGEDORN 1978),

the so-called interphase.


156

Equation 5-1

IR = @oVopR�qr@Ct ∗ 0,46

Equation 5-2

I� = @o$s$R�qr@Ct ∗ 0,79

For the determination of the density in poly(ethylene terephthalate) (PET) MELVEGER

(1972) found a good linear correlation of the bandwidth of the C=O stretching vibration

located at 1730 cm-1. This peak could be fitted with a Lorentzian function in the range of

1530 to 1830 cm-1 and a minimum coefficient of determination of R²>0,99. The density

ρ (g/cm³) can be calculated via (Equation 5-3),

Equation 5-3

u = 305 − ∆vo/F209

where Δν1/2 is the full width (cm-1) at half maximum intensity of the peak at 1730 cm-1.

The results are independent of the orientation in the sample.

5.3.5 Oxygen permeation measurement

The oxygen permeance values of the flexible lid with EVOH as barrier layer (for more

information see previous section) were performed using an Ox-Tran instrument (Modern

Controls, Minneapolis, MN, USA), following the DIN 53 380, T3 (gas carrier method)

with measurement conditions at 23°C and 0% relative humidity. The permeability tests

of the inelastic lid with a thin inorganic barrier layer (SiOx) performed with an optode, an

oxygen measuring device based on fluorescence quenching (Fibox 3 LCD trace, Presens-

Precision Sensing GmbH, Regensburg, Germany). Conditions of determination were

23°C and 0% relative humidity. The permeance value of the film samples is given in

cm³/(m² d bar).

5.3.6 Migration studies

Migration analyses were performed to investigate the influence of supercritical gases on

the migration behavior of high pressure treated film packages. For this purpose a high

pressure treated film sample was compared to an untreated reference sample regarding

the specific migration of two antioxidants (Irganox® 1076 and Irgafos® 168).

Additionally, a comparative screening analysis using GC-FID/MS was performed to


157

characterize the migration potential of components from the film materials and the used

adhesives.

For the analysis of the specific migration and the screening analysis, pouches (13,5 cm x

19,5 cm) were sealed from a multilayer film (PE-LD/PA/EVOH/PA/PE-LD; See Pouch

in Table 5-1) and filled with 100 mL of 95% ethanol. The samples were stored upright.

The headspace of the pouches over the food simulant was approximately 100 mL and the

contact surface of the pouches to the food simulant 95% ethanol was around 3 dm².

The gas composition was a mixture of 20% CO2 and 80% N2. Information about the

pressure system and the applied conditions one can be found in Table 5-1. The external

surface of the samples was in direct contact with the pressure transmitting fluid (water).

The packages were in contact with the food simulant for 12 days at 40°C and additionally

2 days at room temperature during shipping. All migration tests were performed in

duplicate by the accredited Fraunhofer Institute for Process Engineering and Packaging

in Freising, Germany. Compliance testing of the film material with EU legislation for

food contact materials, e.g. (Commission Regulation (EU) No 10/2011) was not part of

the investigations.

A comparative survey of the film samples (Reference to HP sample) was accomplished

by cutting 1 dm² of each sample into small pieces and subsequent extraction with 10 mL

of dichloromethane (DCM) for 1 d at 40°C by total immersion. An aliquot of the 95%

ethanol migration solutions was spiked with the internal standard di-tert-butyl-

hydroxyanisol (BHA) and Tinuvin 234 and studied by GC-FID/MS (gas chromatography

flame ionization detector) for the fingerprint components defined in the dichloromethane

extracts. To enhance the detection limit the internal standard was added to the rest of the

migration solutions which were then reduced to approximately 1 ml under a gentle

nitrogen stream and then analyzed by GC-FID.

The gas chromatographic separation of volatile and low volatile components was

performed on a capillary column DB-1 (length 20 m, internal diameter 0.18 mm, film

thickness 0.18 µm) and the following temperature program: 50°C (2 min isothermal) to

340°C. With this method, organic components are in a molecular weight range of about

150 to about 700 Daltons. For identification of the main components the extracts were

assayed by coupling the gas chromatography and mass spectrometry. The used GC/MS

system was ThermoFinnigan SSQ, column: DB-1-MS - 30 m – 0,25 mm i.d. – 0,25 µm


158

film thickness, temperature program: 80°C (2 min), heating rate 10°C/min, 340°C (30

min), full scan, m/z 40-800. The identification of the obtained spectra was carried out by

comparing spectra of the NIST library and have not been verified by measuring the pure

substances.

Furthermore, the specific migration of two ultraviolet stabilizers was examined.

According to the European Plastics Regulation (EU) No 10/2011 the specific migration

limit of octadecyl-3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate (CAS: 2082-79-3,

trade name e.g. Irganox® 1076) is 6 mg/kg food(-simulant). Tris(2,4-di-tert-

butylphenyl)phosphite (CAS: 31570-04-4; trade name e.g. Irgafos® 168) including its

oxidized form (oxid. Irgafos® 168) is approved as additive for plastics for food contact

applications without a specific restriction. These substances were determined with HPLC

(high-performance liquid chromatography). The chromatographic separation was

performed with a Sphereclone ODS as stationary phase and with 95% ethanol as mobile

phase. The subsequent detection by UV was carried out at a wavelength of 230 nm. The

quantification was performed using an external calibration. Additionally, a standard

addition procedure was performed for the verification of the results.

5.4 Results and Discussion

5.4.1 Crystallinity and density changes of monolayer materials

Apparently, the PE-HD film seems to be unaffected by HPP (Figure 5-5). A tendency of

increasing of the amorphous phase in one type low density polyethylene (PE-LD 1) is

noticeable, but small (Figure 5-6). The meaning of the used statistical box plot

presentation of the data is displayed in Figure 5-4. Slight increasing crystallinity and

decreasing amorphous phase with increasing pressure can be detected for the PE-LD 2

sample treated at 40°C (Figure 5-7).This might indicate a temperature effect. Another PE-

LD polymer (PE-LD 2, not shown here) does not reveal any morphological changes after

a 60 min treatment at 600 MPa at room temperature. Overall, the examination of the

crystalline and amorphous phase in single layer polyethylene does not show significant

changes due to high pressure, even after several days of treatment.

The investigations of the density of PET reveal a slight decrease for different PET

samples. Remarkable is that all these results are of very little order of magnitude (e.g.


159

changes in the third decimal place for PET density) and therefore no influence on polymer

properties is to be expected (Figure 5-8).

Confirming conclusions were made by SANSONE ET AL. (2014). In this study PE-LLD was

tested at varying processing conditions (pressure up to 700 MPa and temperature up to

115°C) and some alterations were found but a negligible effect on the functional

properties. YOO ET AL. (2009) noticed increasing crystallinity of single layer PE-LD, but

in pouches filled with 95% ethanol, which will also have an influence on the results. The

review of FLECKENSTEIN ET AL. (2014) lists previous studies which also dealt with

pressure induced morphological changes in polymers and came to diverging conclusions.

As there are many different ways to determine the crystallinity or density of a sample

(among others Raman spectroscopy, DSC, XRD, FT-IR) it is important to make reference

measurements with the same method for comparable findings. Therefore all quantitative

studies of morphology in this paper were performed using the same methodology with

confocal Raman microscopy.

Figure 5-4. Statistical presentation of the data (box plots)


160

Figure 5-5 Crystalline and amorphous fraction of PE-HD at 23°C (1-5 days) depending on pressure (condition 1A, Table 5-1))

Figure 5-6 Crystalline and amorphous fraction of PE-LD (1) at 23°C (1-5 days) depending on pressure (condition 1A, Table 5-1))

Figure 5-7 Crystalline and amorphous fraction of PE-LD (1) at 40°C (1-5 days) depending on pressure (condition 1B, Table 5-1))


161

Figure 5-8 Changes of density in PET (1) (left) and APET (2) (right); (condition 2C/D and 2D/E, Table 5-1))

5.4.2 Multilayer films in model packages and real packages

5.4.2.1 Influence of layer thickness

No influence of varying layer thickness to the pressure induced morphology changes

could be observed. Figure 5-9 displays the crystallinity of two PE layers (PE-LLD and

PE-MD; part of the model tray, pressure condition 2D, see Table 5-1) in a thermoformed

tray before and after HPP (600 MPa, 5 min). The crystallinity has a general trend to

increase due to high pressure processing, but the results have a low degree of significance.

As similar observations were not made for the single layer films pressurized in water, this

effect will be probably generated by the additional adiabatic heating created by the

compression of the gaseous headspace. Adiabatic heating depends strongly on the kind

of material. PATAZCA ET AL. (2007) tested the quasi-adiabatic heating of several food

products subjected to HPP and observed a temperature increase of about 9°C per 100 MPa

for fatty products like oil, whereas the heating effect for water based products (like skim

milk) was much lower (3 to 4°C/100MPa). KNOERZER ET AL. (2010) tested the

compression heating of polymers (PE-HD, PP, PTFE) at varying initial temperatures (5

to 90°C) up to 750 MPa. They found that the heating rate of PE-HD is much higher than

the rate of water. And the compression heating of gases is likely to be much higher (TING,

BALASUBRAMANIAM 2002). Moreover, it has to be taken into account that during pressure

build up, holding time and pressure release, the temperature distribution in the film and

the layers is not uniform, which explains the scattering of the data. This results agree with


162

studies of ARDIA ET AL. (2004), where a temperature gradient between the center and the

boundary of food products could be shown.

The results obtained for an industrial tray packaging (PET/PEcoex) also indicate local

thermal effects, to be seen in Figure 5-10. As an effect of HPP on the total package, the

crystallinity and measured thickness of three coextruded PE layers has been changed in

both directions (decreasing and increasing). It is evident that, in this case, layer thickness

and morphology is indeed influenced by HPP, but in a non-predictable way.

Figure 5-9 Crystallinity changes in different PE-layers (PE-LLD, PE-MD) of a tray, shown in dependence of layer thickness (Model tray, condition 2D, Table 5-1)


163

Figure 5-10 Scattering of crystallinity and layer thickness in PET/PEcoex tray after HPP of three different PE types coextruded in one tray (Industrial tray, condition 3F, see Table 5-1)

5.4.2.2 Influence of headspace volume and gas composition

The polyethylene as part of an inelastic lid (PET SiOx /Adh./PE-LD) can be regarded as

representative for our findings about the influence of headspace volume and gas

composition on polymer morphology. Figure 5-11 demonstrates that changes of the gas

composition have no effect on the film. The volume fraction of the gaseous headspace

has no effect on the crystallinity, whereas the presence of more gas in the headspace

apparently increases the amorphous content. The effect, however, is minimal and will

probably have no impact on the final polymer function.

Figure 5-11 Crystalline and amorphous content of an inelastic lid (PET SiOx /Adh./PE-LD) in dependence of gas composition (100% CO2, 100% N2 and 20%CO2/80% N2) and headspace volume (15, 30 and 50%), (HP condition 2D, see Table 5-1)


164

5.4.3 Oxygen permeation of multilayer lid films

5.4.3.1 EVOH barrier layer

After HPP of the related model packages, the oxygen permeability of a multilayer lid with

EVOH barrier (Elastic lid, HP condition 2D, see Table 5-1) shows a slight but not

significant increase in permeability. The oxygen permeation of the elastic lid was also

tested by Mondi Technologies (Gronau, Germany), a company of the project consortium.

No significant alterations of the barrier properties 1, 7, 14 or 21 days after the high

pressure treatment could be observed (unpublished results). This is in line with findings

of several other authors: LÓPEZ-RUBIO ET AL. (2005) tested the oxygen barrier of different

PP/EVOH/PP multilayer films at different pressures, exposure times and temperatures,

but without gaseous headspace. They found no significant change of the oxygen

permeability due to HPP. Similar results were obtained by MASUDA ET AL. (1992) and

LARGETEAU (2010). It can be concluded that headspace in packages has (additional to

high pressure) only a small influence on the oxygen barrier properties of multilayer films

based on EVOH.

5.4.3.2 Inorganic barrier layer

The examination of the permeability of an inelastic lid with an inorganic barrier layer

(oPET SiOx/Adh./PE-LD) indicates that at least the inorganic barrier layer is severely

damaged. The permeability increases by a factor of ten, i.e. from 8,8 cm³(STP)/(m² d bar)

to around 80 cm³(STP)/(m² d bar) which is roughly the permeability of the oPET film

alone. (Figure 5-12). Material-specific Raman scans in the film plane at the location of

the interface between the adhesive and the SiOx layer and perpendicular to it (right side

Figure 5-12) give more insight into the failure mechanism: They do not only indicate the

formation of cracks in the SiOx layer, but also an intermixing between PE-LD and the

adhesive and even a distortion of the PET film at the interface.

In previous studies on films with inorganic layers (without headspace), similar results

were obtained: GALOTTO ET AL. (2008) showed that the SiOx layer within a PP/SiOx

multilayer is negatively affected by HP treatment at 400 MPa. Also CANER ET AL. (2000)

tested polymer films with a thin Al metal layer and detected differences in permeability

after HPP when compared with their untreated reference. Only MASUDA ET AL. (1992)

could not find an effect on the oxygen barrier of a PET SiOx/PP laminate subjected to HP


165

treatment at 800 MPa for 2 min at 25°C. Probably, the presence of a small gaseous

headspace is enough to create severe effects. Moreover, especially in the case of rigid

trays the lid must be flexible enough to compensate the compression of headspace gases.

It can be concluded that multilayer films with thin inorganic barrier layers are not suitable

for high pressure processing, especially for MAP.

Figure 5-12 Left: Permeability of lid with inorganic barrier layer (SiOx); Raman scan of destroyed barrier layer (right)

5.4.4 Migration

Comparative screening analyses of dichloromethane extracts and 95% ethanol migration

solutions using GC-FID showed no significant differences for the high pressure treated

film and untreated reference film sample. The identification of fingerprint components of

semi and lower volatility and the semi-quantification results are displayed in Table 5-2.

The higher amount of Caprolactam found in the 95% ethanol migration solutions

compared to the dichloromethane extracts is related to better solubility in 95% ethanol.

Also for the specific migration of Irganox® 1036, Irgafos® 168 and its oxidized form

(oxid. Irgafos® 168) no significant differences between a high pressure treated sample

and its untreated reference were detected (see Table 5-3).


166

Table 5-2. Semi-quantification of fingerprint components in the DCM extracts and the 95% ethanol migration solutions, results given in µg/dm² (as mean values of two determinations)

Substance DCM-Extract Reference [µg/dm²]

DCM-Extract HP [µg/dm²]

Migration Reference [µg/dm²]

Migration HP [µg/dm²]

Caprolactam 499 397 2563 2568 n.i.4 < LD 11 < LD < LD Isomer(s) of di-tert-butylphenol

< LD 5 14 13

Group of chromatographically non-dissolved isomeric or structurally related substituted phenols

152 186 944 903

hexadecanoic acid 26 46 21 < LD octadecanoic acid 40 50 14 15 n.i. < LD < LD 96 112 n.i. < LD < LD 8 19 Erucamide 189 41 472 540 n.i. < LD < LD 9 22 n.i. < LD 28 < LD 14 n.i. < LD < LD 13 25 Irgafos®168 571 453 348 371 Irganox® 1076 und oxid. Irgafos® 168

738 796 362 392

n.i. 1568 1359 209 165 Limit of detection (LD) = 5 µg/dm²

Table 5-3. Specific migration of Irganox® 1076, Irgafos® 168 and oxidized Irgafos® 168

Area-related migration [mg/dm²]

Filling-related migration [mg/kg]5

Sample Reference HP Reference HP

Octadecyl-3-(3,5-di-tert-butyl-4-hydroxyphenyl) propionate (Irganox® 1036)

0,2 0,2 1,2 1,2

Tris(2,4-di-tert-butylphenyl) phosphite (Irgafos® 168)

0,23 0,22 1,4 1,3

Oxid. Tris(2,4-di-tert-butylphenyl)phosphite (oxid. Irgafos® 168)

0,04 0,04 0,24 0,24

Limit of detection (LD) 0,02 0,11

4 Not identified 5 Based on a surface-volume-ratio of 6 dm²/kg (according to the EU cube model)


167

These results stand in good agreement with the studies of SCHAUWECKER ET AL. (2002),

who did not observe migration of 1,2-propanediol (pressure transmitting fluid) into

PET/PA6/Al/PP (146 µm) pouches compared to migration under atmospheric conditions.

LARGETEAU (2010) tested the migration of PA/PE compounds into several food

simulating liquids (water, acetic acid, ethyl alcohol, isooctane). In this study also no effect

of HP treatment on the migration was observed. Similar results were found by MAURICIO-

IGLESIAS ET AL. (2010). Here it could be shown that there is no further extraction of

polymer components or additives due to supercritical fluids like CO2 during high pressure

processing and following storage time. When considering the reported overall reduction

of diffusion coefficients in polymers subjected to high pressure (RICHTER ET AL. 2010)

together with the small amount of CO2 actually present in MAP packaging, it is very

unlikely that an additional extraction of additives will occur in practice.

5.5 Conclusions

The results show small structural changes in relevant packaging polymers subjected to

high pressure processing, which are negligible for the macroscopic polymer properties

and their industrial application. It could especially be concluded that no relevant structural

changes arise from high pressure alone.

The combination of HPP and headspace gases in modified atmosphere packages does not

significantly affect the morphology of the inner layers from polyethylene. Small

alterations like an increase in the crystallinity of PE can be observed. Generally, however,

HPP leads to increasing and decreasing of the crystallinity in the same sample at the same

time at different locations. A reason for this observation might be the generation of local

variations of the temperature due to adiabatic heating inside the packaging.

The composition of the gas in the headspace does not have an observable effect on the

packaging. Only the amount of headspace volume seems to affect the amorphous phase

in polyethylene and the overall integrity of a tray packaging.

The results also confirm the absence of possible additional risks for the consumer by

migration of polymer additives into the product.

The oxygen permeability of non-treated and high pressure treated multilayer films

showed that films on the basis of EVOH layers remained intact, also in combination with


168

headspace gases. Multilayer films with thin inorganic barrier layers are subject to severe

damages due HPP. These damages do not only affect the inorganic layers alone, but also

the adjacent zones of the polymeric layers. The reason for these effects is still not clear.

A simple explanation could be the low elasticity of the inorganic layers which is known

to lead to a complete failure of their barrier properties already at a linear strain of several

percent. Moreover, high pressure gradients in the vicinity of the inorganic barrier layers

might lead to higher rate of formation of gas bubbles. This would mean that extreme local

differences in gas permeability should be generally avoided by a proper design of the

laminate structure of films to be used in HPP.

A visible negative effect of headspace gases are blisters and white spots inside the

packaging material or also inside the food simulant which can be assigned to localized

gas decompression. Here, the results still do not allow for systematic approaches. It is,

however, possible to reduce the risk of the occurrence of these failures by considering

some packaging design aspects: As these defects appear more frequently at weak points

of a packaging, e.g. in the edges and thinnest parts of a tray, risks of damage might be

decreased by reducing the thickness variation of a tray. A more uniform wall thickness

distribution, made possible e.g. by a locally adapted heating regime in the thermoforming

process (CLAUS ET AL. 2013) would be a viable approach. Moreover, weak boundary

layers between coextruded layers or films and adhesives and inhomogeneities inside the

material raise the risk of explosive decompression failures. A zone where the formation

of White Spots was often observed is the transition phase between coextruded

polyethylene layers. So the formation of localized damages is still an important issue for

research in packaging materials for HPP applications.

Overall, however, the use of headspace gases in packages for high pressure treatment is

conceivable, as long as proper material combinations are applied.

5.6 Acknowledgments

Funding of a part of this work by the Bundesministerium für Ernährung, Landwirtschaft

und Verbraucherschutz BMELV, grant-number PGI-0313849A, is gratefully

acknowledged. The authors would like to thank Mondi Technologies (Gronau, Germany)

and the Deutsches Institut für Lebensmitteltechnik (DIL, Quakenbrück, Germany) for the

preparation of the films used in this study and for the great collaboration.


169


ARDIA, A.; KNORR, D.; HEINZ, V. (2004): Adiabatic heat modelling for pressure build-

up during high-pressure treatment in liquid-food processing. Mixing, Heat and Mass

Transfer. In Food and Bioproducts Processing 82 (1), pp. 89–95. DOI:

10.1205/096030804322985362.




10.1002/app.25085.





CANER, C.; HERNANDEZ, R. J.; PASCALL, M. A. (2000): Effect of high-pressure

processing on the permeance of selected high-barrier laminated films. In Packaging

Technology & Science 13 (5), pp. 183–195. DOI: 10.1002/1099-1522(200009)13:5.

CLAUS, R.; STEIN, M.; BACH, S.; MAJSCHAK, J.-P.; PARTSCH, U.; GRIEßMANN, H.

(2013): An innovative contact heating method in the thermoforming process. In :

EUROTEC® 2013. Proceedings of the 2nd European technical conference and

exhibition. Lyon, France. Society of Plastics Engineers: Society’s online technical

library.



pp. 303–310. DOI: 10.1002/pts.858.




10.1002/pts.2018.




170



HOLLRICHER, O.; IBACH, W. (2011): High-resolution optical and confocal microscopy.

In Thomas Dieing, Olaf Hollricher, Jan Toporski (Eds.): Confocal Raman

microscopy. With assistance of W. T. Rhodes. Berlin Heidelberg: Springer (Springer

Series in Optical Sciences, 158), pp. 1–20.



Reviews 2 (4), pp. 274–297.

KNOERZER, K.; BUCKOW, R.; VERSTEEG, C. (2010): Adiabatic compression heating

coefficients for high-pressure processing–a study of some insulating polymer

materials. In Journal of Food Engineering 98 (1), pp. 110–119.



LAX, E.; ANS, J. d.; BLACHNIK, R. (1998): Taschenbuch fur Chemiker und Physiker.

Elemente, anorganische Verbindungen und Materialien, Minerale. Band 3. Berlin

Heidelberg: Springer DE (Elemente, anorganische Verbindungen und Materialien,

Minerale).










MAURICIO-IGLESIAS, M.; JANSANA, S.; PEYRON, S.; GONTARD, N.; GUILLARD, V.

(2010): Effect of high-pressure/temperature (HP/T) treatments of in-package food on

additive migration from conventional and bio-sourced materials. In Food Additives &


171

Contaminants: Part A: Chemistry, Analysis, Control, Exposure & Risk Assessment

27 (1), pp. 118–127. DOI: 10.1080/19440040903268054.

MAXFIELD, J.; STEIN, R. S.; CHEN, M. C. (1978): Polarized Raman studies of crystalline

and amorphous orientation in polyethylene. In J. Polym. Sci. Polym. Phys. Ed. 16 (1),

pp. 37–48. DOI: 10.1002/pol.1978.180160104.

MELVEGER, A. J. (1972): Laser-Raman study of crystallinity changes in poly(ethylene

terephthalate). In J. Polym. Sci. A-2 Polym. Phys. 10 (2), pp. 317–322. DOI:

10.1002/pol.1972.160100211.

PATAZCA, E.; KOUTCHMA, T.; BALASUBRAMANIAM, V. M. (2007): Quasi-adiabatic

temperature increase during high pressure processing of selected foods. In Journal of

Food Engineering 80 (1), pp. 199–205. DOI: 10.1016/j.jfoodeng.2006.05.014.

Commission Regulation (EU) No 10/2011, 1/14/2011: Regulations on plastic materials

and articles intended to come into contact with food.



RICHTER, T.; LANGOWSKI, H.-C. (2005): Hochdruckbehandlung von verpackten

Lebensmitteln - Beurteilung flexibler, polymerer Verpackungsmaterialien für den

Einsatz in der Hochdruckpasteurisation. In Der Weihenstephaner 44 (2), pp. 83–86.



566. DOI: 10.1080/08957959.2010.531722.

SANSONE, L.; ALDI, A.; MUSTO, P.; AMENDOLA, E.; MENSITIERI, G. (2014): Effects of

high pressure treatments on polymeric films for flexible food packaging. In

Packaging Technology & Science.

SCHAUWECKER, A.; BALASUBRAMANIAM, V. M.; SADLER, G.; PASCALL, M. A.;

ADHIKARI, C. (2002): Influence of high-pressure processing on selected polymeric

materials and on the migration of a pressure-transmitting fluid. In Packaging



172

STERR, J. (2012): The effect of high pressure on the morphology of polymers – a Raman

spectroscopic study. 7th International Conference on High Pressure Bioscience and

Biotechnology (HPBB2012). Otsu, 11/2/2012.

STROBL, G. R.; HAGEDORN, W. (1978): Raman spectroscopic method for determining

the crystallinity of polyethylene. In J. Polym. Sci. Polym. Phys. Ed. 16 (7), pp. 1181–

1193. DOI: 10.1002/pol.1978.180160704.

TING, E.; BALASUBRAMANIAM, V. M. (2002): Determining thermal effects in high-

pressure processing. In Food Technology 56 (2), pp. 31–35.

YOO, S.; LEE, J.; HOLLOMAN, C.; PASCALL, M. A. (2009): The effect of high pressure

processing on the morphology of polyethylene films tested by differential scanning

calorimetry and X-ray diffraction and its influence on the permeability of the

polymer. In J. Appl. Polym. Sci. 112 (1), pp. 107–113. DOI: 10.1002/app.29401.

DISCUSSION AND CONCLUSION

173

6 Discussion and Conclusion

High pressure processing of food in modified atmosphere packaging is part of a hurdle

concept for prolonging the shelf life and for improving the quality of temperature

sensitive foods. As common gases are in a supercritical state during pressure treatment,

special attention must be paid to interactions between the gases and materials. The

packaging integrity may be negatively affected. The aim of this study was to identify

failure mechanisms and to carry out experiments to acquire relevant data. The following

sections discuss and summarise the findings. An overview of the key results is shown in

Figure 6-1.

6.1 Influence of high pressure processing of packages with modified

atmospheres on packaging integrity

6.1.1 Reversible structural changes

In order to understand the effects on the polymeric packaging, the mechanisms during

high pressure treatment must first be understood. Up until now, no in-situ measurements

of gas permeation under hydrostatic pressures of up to 2,000 bars have been carried out.

The development of a new measurement method based on fluorescence quenching of

oxygen has enabled the quantification of oxygen permeation processes in polymers

(Chapter 4).

The results of the permeation measurements indicated a decrease in permeability by up

to a factor of 70. This was comparable to results concerning the transport processes of

aromatic compounds under similar pressure conditions (RICHTER ET AL. 2010). The

reduction in permeability was put down to the fact that the free volume of the polymer is

markedly reduced under hydrostatic pressure. However, simulation studies of Sarrasin

predicted a decrease in the amorphous volume of polyethylene by only a factor of 1.06 at

pressures up to 2,000 bars (SARRASIN ET AL. 2015). Hence the detected strong reduction

of the oxygen permeability cannot be explained by merely this factor. An explanation

might by the marked decrease in chain motion at high pressures which is necessary for

the fluctuation of voids in the matrix and therefore responsible for the movement of

diffusing molecules. The chains are arranged in a quasi-crystalline state during pressure


174

treatment which diminishes the permeation of gases in polymers. Related comments were

made by Gorbachev who blamed “increasing intermolecular interaction, decreasing chain

flexibility and increasing packing density of the macromolecules” for the reduced

permeability of gases in polymers (GORBACHEV ET AL. 1977).

The results provide an important contribution for understanding the processes taking

place under pressure, even though the determination of the solubility and diffusion

coefficients is not yet possible. As the breakthrough time of oxygen in a polyethylene

film was faster than the pressure build-up time (about 2 minutes), those coefficients could

not be determined using the ‘lag time’ method. The breakthrough time of oxygen in a

polyethylene film of 100 µm thickness is about 55 seconds under ambient pressure and

temperature conditions, as calculated using an averaged diffusion coefficient in PE

(MOISAN 1985). As a decrease both in diffusion and solubility coefficient was observed

for other substances (RICHTER, STERR ET AL. 2010), this is likely to have happened also

in the case of oxygen.

Future work should use polymers with lower diffusion coefficients to determine the gas

solubility under high hydrostatic pressures. Also, the measurement of the permeability of

other gases such as CO2 with optical sensors is desirable.

6.1.2 Irreversible changes

The review in Chapter 3 summarises the irreversible structural changes in polymers due

to pressure application. Knowledge of these changes is of great importance for predicting

the suitability of a polymer for HPP as the structure, namely the crystallinity and density,

influence the barrier and mechanical properties. In the present work, structural changes

were determined via Raman spectroscopy (Chapter 5.3.4). The influence of headspace

volume and composition, film thickness, initial polymer density, maximum applied

pressure and treatment time was investigated. The results clearly show that the changes

in crystallinity are not uniform or homogeneous throughout the polymeric matrix but

seem to be dependent on thermal gradients. The pressure range, pressure holding time

and temperature (under moderate temperature conditions) did not significantly influence

the crystalline and amorphous phases of the polymer. In contrast, an increase in headspace

volume increased the amorphous content of polyethylene. A possible reason for this is

the increased amount of dissolved gas in the polymer and the concomitant plasticising of

the intermediate phase. The crystalline phase did not undergo any significant


175

transformation, possibly indicating that the structure of the polymer crystals is in general

insensitive to hydrostatic pressures of up to 6,000 bars.

In summary, none of the directly and indirectly induced changes in polymer morphology

which were detected will play a crucial role for industrial application of HPP as the

changes were small in absolute terms.

6.1.3 Changes in oxygen permeability

The oxygen permeation measurements after high pressure treatment on pouches and tray

packaging with headspace (Chapter 5.4.3) showed good agreement with results from

literature for experiments on vacuum packaging (LARGETEAU 2010; LÓPEZ-RUBIO ET AL.

2005; MASUDA ET AL. 1992). In these studies ethylene - vinyl alcohol copolymer (EVOH)

was used as a barrier layer. Even in the presence of supercritical gases, the barrier

properties of EVOH did not show any significant changes after HPP. This barrier material

can therefore be particularly recommended for HPP. Only if there is delamination or

blister formation, the oxygen transmission will increase significantly. Also, inorganic

barrier layers were found to be sensitive to blistering as well as to mechanical stress under

HPP conditions. The result is a total loss of barrier properties in most cases. These

materials should therefore be avoided for the use in modified atmosphere packaging with

HPP.

6.1.4 Migration

No significant additional migration (global or specific) due to a possible extraction of

substances with supercritical carbon dioxide (scrCO2) was detected (Chapter 5.4.4). This

was expected due to the small amount of CO2 present in the packaging. For common

extraction processes such as used for decaffeinating coffee, higher amounts of gases are

necessary. The findings agree with results in the literature on the global migration of

packaging constituents into food simulating liquids after high pressure treatment of

vacuum packaging (AYVAZ ET AL. 2016; JULIANO ET AL. 2010). The findings also agree

with experiments on the extraction of plasticisers out of polyamide (PA11) using CO2 at

pressures of up to 40 bars and temperatures of up to 120°C (FLACONNÈCHE ET AL. 2001).

It can be concluded that if a polymer is in conformity with the migration limits under

atmospheric conditions as specified by regulations and if the polymer remains undamaged

under HPP treatment, then the requirements will also be met under HPP conditions.


176

An overview of the results of this work is given in Figure 6-1.

Figure 6-1 Overview of the results of the present study

6.2 General guidelines for packaging design and materials used for

modified atmosphere packaging for high pressure processing

The following section is based on the experimental data, the outcomes of the discussion

and information available in the literature. The information can be put forward as general

guidelines for processing conditions, material properties and packaging design for

practical use.

6.2.1 Material properties

The heterogeneity of a material was identified as being one of the parameters having the

highest impact on decompression failure. Inclusions in the polymer, such as additives and

other components, may act as nucleation points for bubble formation and may result in

severe damage in most cases (Chapter 2.4.2.2). By way of example, two pouches with


177

multilayer films of comparable composition are presented in Figure 6-2. The material on

the left had inhomogeneities which acted as nucleation points for bubbles, whereas the

material on the right was additive free. The materials show clearly differences in the

amount of damage.

Figure 6-2 Pouches with inhomogeneities (left) and without inhomogeneities (right) and bubble formation after HPP. Own unpublished results from investigations made for industrial clients.

The packaging materials have to be flexible enough to take up the pressure and to

reversibly withstand the compression of the headspace and product. The mechanical

properties of some rigid materials such as inorganic barrier layers cannot withstand the

volume decrease. Irreversible macroscopic damage, for example delamination and flex-

cracks, may result. When using rigid materials for tray packaging it is important that at

least the lid of the packaging is able to take up the pressure and compensate the volume

reduction. Also, packaging components with porous and foam-like structures such as

cardboard, paper and expanded polystyrene turned out to be unsuitable for high pressure

processing (Figure 6-3 used for illustration) because the voids are irreversibly compressed

and are not restored after decompression (CANER ET AL. 2004; OCHIAI, NAKAGAWA 1992).

In Chapter 2, Section 2.4.2 it was shown that factors such as higher molecular weight,

higher crystalline fraction, lower elasticity and lower viscosity of the polymers are able

to reduce or even prevent the formation of bubbles and blisters.

6.2.2 Packaging geometry and design

The packaging design is important for avoiding decompression failure and damage to the

packaging integrity. For example, tray packaging should have large angles in the curved

zone between the walls and the base. The profile of an embossing should be avoided. In

general the tray and the packaging film should be flat, preventing gas pockets between


178

the material and the food product (see Figure 6-3 exemplary for an unsuitable design).

High temperature gradients due to the high heat of compression of the gases are then

prevented in these positions (Chapter 5.5). In general, the headspace volume should be as

small as possible. Inlays made of cardboard turned out to be unsuitable for the high

pressure process (see Figure 6-3, left picture of a cardboard inlay after HPP).

Considering the structure of a multilayer film, it should be kept in mind that widely

differing properties of adjacent polymers or materials, such as differences in their

mechanical or barrier properties, may lead to considerable gas pressure gradients and

hence to bubble formation or delamination (Chapter 5.5).

Figure 6-3 Left: Cardboard inlay after HPP. Right: Example of an unsuitable packaging design. Own unpublished results from investigations made for industrial clients.

6.2.3 Processing conditions

The process conditions such as the compression and decompression rates and the process

temperature also have a considerable influence. It is recognised that the pressure build-

up rate should be as high as possible and the pressure release rate as low as possible in

order to get optimum results (Chapter 2.4.2.3). The principle of stepwise and slow

decompression has already been realised in industrial applications by MULTIVAC

GmbH, Wolfertschwenden (RICHTER 2014). The treatment time also has an influence on

polymer-gas combinations with long penetration time. If the time is prolonged more gas

is able to dissolve in the polymer and the result is more blisters on decompression.

Similarly, an increase in the maximum applied pressure increases the amount of gas

which is able to dissolve in the materials (Chapter 2.4.2.3). Anyhow a minimum treatment

time combined with a minimum applied pressure is necessary for microbial inactivation

and should not fall below the critical values.


179

A summary of the parameters and properties which have to be taken into account for high

pressure processing is given in Table 6-1.

Table 6-1 Summary of the parameters and properties which should be taken into account for high pressure processing. Results achieved within this thesis are indicated in bold.

Material properties

Dos Don’ts

• Flexibility of the film • High in molecular weight • High in crystalline fraction

• Rigid materials • Foam-like structures (cardboard,

expanded polystyrene) • Inorganic barrier layers • Inhomogeneities inside the

polymers

Packaging design parameters

Dos Don’ts

• Large angles • Flat tray bottom • Low headspace volume

• Gas pockets between material and food product

• Embossing • Layer structure: o Strong pressure gradients at

barrier layers o Large differences in mechanical

properties

Process parameters

Dos Don’ts

• High pressure build-up rate • Slow pressure decrease • Stepwise pressure decrease • Moderate process temperatures

• Spontaneous pressure decrease • High process temperatures • Excessive high pressures (keep the

pressure to the minimum required) • Excessive treatment times (keep the

treatment time to the minimum required)


180

6.3 Final remarks and outlook

It has been demonstrated that commonly used flexible polymeric materials are generally

suitable for modified atmosphere packaging for high pressure processing at moderate

temperatures. However, some design parameters and material properties must be taken

into account when using modified atmosphere packaging for HPP. The fulfilment of

standard regulatory requirements by the material with respect to its migration properties

should be verified for each case individually.

It was found that explosive decompression failure is a phenomenon influenced by a host

of (combined) criteria. Only a limited number of parameters are known to reduce or

prevent the occurrence of bubbles and blisters. For better prediction and prevention of

decompression failure, further research is necessary focussing on the process parameters

applied for HPP. Additional studies are also needed on the high pressure treatment of

modified atmosphere packaging at elevated temperatures.

The application of nanotechnology, active packaging and/or packaging with absorbing

additives, such as zeolites, should be considered for future work. The in-situ measurement

method (Chapter 4) which was developed will facilitate understanding the chemical and

physical reactions which take place under pressure. Similarly, analysis of CO2 reactions

and CO2 transport mechanisms is also conceivable using the recently developed optical

device of PreSens GmbH (APOSTOLIDIS, A. 2017) for measuring CO2 concentrations in

gaseous and liquid phases.


181

6.4 Chapter bibliography

APOSTOLIDIS, A. (2017): Non-invasive online CO2 measurement with chemical optical

CO2 sensors. Chemical optical CO2 sensors provide a versatile tool for online

monitoring of dissolved CO2 in various scale and designs.

https://www.presens.de/knowledge/publications/application-note/non-invasive-

online-co2-measurement-with-chemical-optical-co2-sensors-578.html. Accessed 07

May 2017.

AYVAZ, H.; BALASUBRAMANIAM, V. M.; KOUTCHMA, T. (2016): High pressure effects

on packaging materials. In V. M. Balasubramaniam, Gustavo V. Barbosa-Cánovas,




CANER, C.; HERNANDEZ, R. J.; HARTE, B. R. (2004): High-pressure processing effects

on the mechanical, barrier and mass transfer properties of food packaging flexible

structures: a critical review. In Packaging Technology & Science 17 (1), pp. 23–29.

DOI: 10.1002/pts.635.

FLACONNÈCHE, B.; MARTIN, J.; KLOPFFER, M.-H. (2001): Permeability, diffusion and

solubility of gases in polythylene, polyamide 11 and poly(vinylidene fluoride). In Oil

Gas Sci. Technol. – Rev. IFP Energies nouvelles 56 (3), pp. 261–278.

GORBACHEV, B. N.; NIKIFOROV, N. N.; CHEBANOV, V. M. (1977): Air permeability of

polymers subjected to high pressures. In Polymer Mechanics 12 (6), pp. 969–972.

DOI: 10.1007/BF00856503.



Reviews 2 (4), pp. 274–297.






182








MOISAN, J. Y. (1985): Effects of oxygen permeation and stabiliser migration on

polymer degradation. In J. Comyn (Ed.): Polymer Permeability. Dordrecht: Springer

Netherlands, pp. 119–176.





519.



5/15/2014. Patent no. US8771773B2.



566. DOI: 10.1080/08957959.2010.531722.

SARRASIN, F.; MEMARI, P.; KLOPFFER, M.-H.; LACHET, V.; TARAVEL CONDAT, C.;

ROUSSEAU, B.; ESPUCHE, E. (2015): Influence of high pressures on CH4, CO2 and

H2S solubility in polyethylene. Experimental and molecular simulation approaches

for pure gas and gas mixtures. Modelling of the sorption isotherms. In Journal of

Membrane Science 490, pp. 380–388. DOI: 10.1016/j.memsci.2015.04.040.

SUMMARY

183

7 Summary

Over the past twenty years high pressure processing (HPP) has become one of the most

important alternative industrial preservation methods to the thermal pasteurization of food

products. Pressures up to 6,000 bars and moderate temperatures inactivate vegetative

cells, yeast and mould, whereas temperature-sensitive food ingredients remain almost

unaffected. This method can therefore be used to give consumers fresh products with

fewer preservatives.

In industrial applications the food is packaged prior to high pressure processing in order

to avoid recontamination. In most cases, vacuum packaging systems are used. As part of

a hurdle concept, modified atmosphere packaging has become ever more common in

recent years for applications such as colour stabilisation of fresh meat and the inhibition

of microorganism growth by carbon dioxide. Improved product presentation is an

additional positive side-effect. The gases commonly used for food packaging are

nitrogen, oxygen and carbon dioxide.

Numerous research studies have been undertaken to examine the influence of high

hydrostatic pressure on food products, ingredients and microbial species. The impact of

pressure on vacuum packaging materials has previously been investigated. However, up

to now very few studies have been undertaken on the influence of headspace gases on the

integrity of polymers under high pressure.

The resulting effects on packaging under HPP were differentiated into direct and indirect

effects, where direct effects are induced by high hydrostatic pressure alone. Indirect

effects arise from compression of the polymeric materials or food products and primarily

due to compressed gases in the headspace of the packaging. It became clear that both

direct effects and indirect effects can cause reversible and irreversible changes to the

polymer structure. However, no clear trends could be found in the scientific data. The

reversible structural changes induce reversible changes to the transport properties of the

gases in polymers and are discussed in the second part of this work.

Irreversible structural changes can influence the barrier properties as well as the

mechanical properties of the packaging. Raman spectroscopic studies were performed to

analyse the density and the crystalline and amorphous parts of relevant polymers such as

SUMMARY

184

polyethylene and polyethylene terephthalate. Key features of this method are non-

destructive measurement at very high spatial resolution. The resulting data reveals small

but technically not relevant changes to the polymer structure dependent on factors such

as the pressure holding time, film thickness, headspace volume and gas composition.

Highly inhomogeneous temperature distributions in the packaging due to different

adiabatic heating rates of materials, gases or food product constituents under pressure

may be an explanation for the inhomogeneous structural changes in the polymer part of

tray packaging. However, these small changes were deemed to have a negligible influence

on the mechanical or barrier properties for industrial applications.

The oxygen permeability of organic barrier layers (e.g. ethylene-vinyl alcohol copolymer)

and inorganic barrier layers (e.g. silicon dioxide) was measured under high pressure

conditions. The former were found to be unaffected by HPP whereas the inorganic barrier

layers were found to be not stable enough for high pressure applications.

Modified atmosphere packaging (MAP) for HPP must also fulfil all legal requirements

concerning the migration of packaging components into foodstuffs, including the

additional extraction of constituents by supercritical carbon dioxide. No increase in the

specific or overall migration was measured in these studies.

The second part of this thesis deals with the phenomenon of bubble formation in modified

atmosphere packaging upon high pressure treatment. Although the development of

bubbles and blisters and the delamination of multilayer films, including collapse of the

overall packaging integrity, have been reported in previous studies, the factors

influencing these effects are not totally clear. The objective of this work was therefore to

first of all identify relevant factors causing bubble formation in polymers under HPP

conditions.

The occurrence of blisters is also known as explosive decompression failure (XDF) in

other areas such as in the petroleum industry. It is due to the outgassing of polymers upon

rapid decompression, caused by the rapid change in the thermodynamic equilibrium and

the subsequent formation of bubbles in oversaturated materials. It was clear from the

literature that the solubility of gases under high pressure increases significantly. This

plays an important role because an increased amount of gas dissolves in polymers under

high pressure. Besides the solubility, the diffusion and permeation coefficients were also

found to influence bubble formation, but no relevant scientific data about the

SUMMARY

185

measurement of gas transport under high hydrostatic pressure could be found. Only

dynamic permeation measurements under moderate partial pressure differences had

previously been performed.

The objective of this part of the work was therefore to develop a new measurement

method based on the principle of dynamic fluorescence quenching. This allows the in-

situ measurement of the oxygen concentration and oxygen permeation under hydrostatic

pressures of up to 2,000 bars. It was demonstrated that the permeation coefficient of

oxygen in a polyethylene film is reduced by a factor of between 35 and 70 at 2,000 bars

applied pressure compared to the value under atmospheric conditions. The results are in

good agreement with findings on the permeation of aromatic compounds through

polymers under similar conditions. A reduction in the polymeric free volume and

diminished chain motion were deemed to be the cause.

As it was not possible to determine the solubility and diffusion coefficient for gases in

polyethylene under high pressure, future work should further develop the method to allow

analysis of various other polymers. Also, modifying the experimental setup with an

optical sensor system for the detection of carbon dioxide concentration is an option.

Simultaneous in-situ tracking of O2 and CO2 permeation would then be possible.

This work showed that commonly used flexible polymers are suitable for the high

pressure treatment of vacuum packaging and modified atmosphere packaging at moderate

temperatures. Regulatory requirements are still fulfilled after HPP, provided the

packaging meets requirements under atmospheric conditions. However, some key design

features concerning the packaging and film materials should be addressed for industrial

applications. These include the avoidance of thin inorganic barrier layers or foamed and

rigid materials.

In conclusion it can be stated that this work contributes to the safe use of vacuum

packaging and modified atmosphere packaging for the high pressure treatment of food.

186

ZUSAMMENFASSUNG

187

8 Zusammenfassung

In den letzten zwanzig Jahren hat sich die Hochdruckbehandlung (HPP) zu einer der

wichtigsten alternativen Konservierungsmethoden für die thermische Pasteurisierung von

Nahrungsmittel in der industriellen Anwendung entwickelt. Bei Drücken bis zu 6.000 bar

und moderaten Temperaturen werden vegetative Zellen, Hefen und Schimmelpilze

inaktiviert, während wertvolle temperaturempfindliche Lebensmittelinhaltsstoffe so gut

wie unverändert bleiben. Diese Methode ermöglicht es dem Handel, frische Produkte mit

weniger Konservierungsstoffen anzubieten.

Für industrielle Anwendungen wird das Lebensmittel vor der Hochdruckbehandlung

verpackt, um eine Rekontamination zu vermeiden. Bei den meisten Anwendungen

werden Vakuumverpackungen verwendet. In den vergangenen Jahren jedoch wurden im

Hochdruckprozess immer häufiger Verpackungen mit modifizierter Atmosphäre im

Kopfraum (MAP) eingesetzt, da es zusätzliche Vorteile bietet, wie die Farbstabilisierung

von frischem Fleisch oder die Hemmung des Mikroorganismenwachstums durch

Kohlendioxid. Gase, die für Lebensmittelverpackungen (MAP) am häufigsten eingesetzt

werden, sind Stickstoff, Sauerstoff und Kohlendioxid.

Zahlreiche Untersuchungen wurden durchgeführt, um den Einfluss von hohem

hydrostatischem Druck auf verschiedene Lebensmittel, Inhaltsstoffe und

Mikroorganismen zu untersuchen. Auch die Auswirkungen auf Verpackungsmaterialien

als Teil von Vakuumverpackungen wurden untersucht. Aber nur sehr wenige

wissenschaftliche Arbeiten betrachteten den Einfluss von Gasen im Kopfraum der

Verpackung auf die Integrität von Polymeren bei der Hochdruckbehandlung.

In dieser Arbeit wurden daher, auf Basis relevanter wissenschaftlicher Arbeiten, mögliche

Auswirkungen von HPP auf die Verpackung in verschiedene Effekte eingeteilt. Dabei

wurde zwischen direkten und indirekten Effekten unterschieden. Direkte Effekte werden

durch hohen hydrostatischen Druck allein induziert und indirekte Effekte durch

komprimierte Kunststoffe oder Lebensmittel, vor allem aber durch komprimierte Gase

im Kopfraum der Verpackung. Es wurde deutlich, dass sowohl direkte als auch indirekte

Effekte die Polymerstruktur reversibel und irreversibel verändern können. Ein eindeutiger

Trend konnte jedoch nicht beobachtet werden. Die reversiblen Änderungen der

ZUSAMMENFASSUNG

188

Polymerstruktur spiegeln sich in reversiblen Veränderungen der Transporteigenschaften

von Gasen in Polymeren wieder und werden im zweiten Teil dieser Arbeit diskutiert.

Irreversible Strukturveränderungen können sowohl die Barriereeigenschaften als auch die

mechanischen Eigenschaften der Verpackung beeinflussen. Raman-spektroskopische

Untersuchungen wurden durchgeführt, um die Dichte sowie den kristallinen und

amorphen Teil relevanter Polymere quantitativ zu analysieren. Die Vorteile dieser

Methode waren die zerstörungsfreie Messung bei gleichzeitig sehr hoher räumlicher

Auflösung. Die resultierenden Daten zeigen kleine, aber technisch nicht relevante

Änderungen der Polymerstruktur in Abhängigkeit von Faktoren wie der Druckhaltezeit,

der Schichtdicke, dem Kopfraumvolumen und der Gaszusammensetzung. Es wird

erwartet, dass insbesondere inhomogene Temperaturverteilungen in der Verpackung

aufgrund der unterschiedlichen adiabaten Erwärmung von Materialien, Gasen und/oder

Produktbestandteilen eine Erklärung für inhomogene Strukturveränderungen in

Polymeren von Tray-Verpackungen sein können. Eine Relevanz dieser minimalen

Änderungen für die mechanischen Eigenschaften oder der Barrierefunktion bei der

industriellen Anwendung konnte nicht beobachtet werden.

Die Sauerstoffpermeabilität von organischen sowie von anorganischen Barriereschichten

wurde untersucht. Erstere zeigten sich von HPP nicht beeinflusst, wohingegen sich

anorganische Sperrschichten für Hochdruckanwendungen als unzureichend

herausstellten. Die mechanischen Eigenschaften sind nicht ausreichend um eine

Kompression (insbesondere des Kopfraums) auszugleichen. Außerdem entsteht an diesen

Schichten ein hoher Partialdruckgradient im Vergleich zu benachbarten Polymeren, der

die Bildung von Blasen forciert.

Eine zusätzliche Extraktion von Verpackungskomponenten in Lebensmittel durch

überkritisches Kohlendioxid konnte in dieser Arbeit nicht nachgewiesen werden. Die

Untersuchungen ergaben keine erhöhten Werte der spezifischen Migration oder der

Gesamtmigration während und nach der Hochdruckbehandlung.

Der zweite Teil dieser Arbeit beschäftigt sich mit dem Phänomen der Blasenbildung in

MAP nach der Hochdruckbehandlung. Obwohl die Entwicklung von Blasen und

Delaminationen von mehrschichtigen Filmen bereits in früheren Studien diskutiert

wurden, konnten die beeinflussenden Faktoren nicht vollständig geklärt werden. Ziel

ZUSAMMENFASSUNG

189

dieser Studie war es zunächst, relevante Faktoren für die Blasenbildung in Polymeren bei

HPP zu identifizieren.

Das Auftreten von Blasen ist in anderen Anwendungsgebieten wie der Erdölindustrie als

"explosive decompression failure" (XDF) bekannt. Das Phänomen wird durch die

Übersättigung von den in Polymeren gelösten Gasen erklärt. Die schnelle Veränderung

des thermodynamischen Gleichgewichts nach rascher Dekompression ermöglicht die

Bildung von Blasen in übersättigten Materialien. Aus der Literatur wurde deutlich, dass

die Löslichkeit von Gasen bei hohen Drücken deutlich zunimmt. Neben der Löslichkeit

wurde auch der Diffusions- und Permeationskoeffizient, also die Transporteigenschaften

von Gasen in Polymeren unter hohen Drücken als Einflussfaktoren für die Blasenbildung

genannt. Allerdings existierten keine relevanten wissenschaftlichen Daten zur

Bestimmung der Gaspermeation bei hohem hydrostatischem Druck.

Ziel dieses Teils der Arbeit war es daher, ein neues Verfahren zu entwickeln, das auf dem

Prinzip der dynamischen Fluoreszenzlöschung basiert und die in situ Messung der

Sauerstoffkonzentration und Sauerstoffpermeation bei hydrostatischen Drücken bis zu

2.000 bar ermöglicht. Es konnte gezeigt werden, dass, verglichen mit Werten bei

atmosphärischen Bedingungen, der Permeationskoeffizient von Sauerstoff in Polyethylen

um einen Faktor zwischen 35 und 70 bei 2.000 bar verringert wird. Eine Verringerung

des polymeren freien Volumens und eine verminderte Kettenbewegung werden als ein

Grund für diese Ergebnisse gesehen.

In dieser Arbeit konnte gezeigt werden, dass handelsübliche flexible Polymere

grundsätzlich für die Hochdruckbehandlung von Vakuumverpackungen sowie für

Verpackungen mit modifizierter Atmosphäre bei moderaten Temperaturen geeignet sind.

Unter Annahme, dass die verwendeten Materialien bei atmosphärischen Bedingungen

den gesetzlichen Anforderungen genügen, kann davon ausgegangen werden, dass die

Anforderungen auch nach der Hochdruckbehandlung erfüllt sein werden. Allerdings

sollten einige entscheidende Parameter für die Verpackungen und Folienmaterialien bei

der industriellen Anwendung berücksichtigt werden.

Zusammenfassend kann diese Arbeit zur sicheren Anwendung von polymeren

Packstoffen für die Hochdruckbehandlung von Lebensmitteln beitragen.

190

CURRICULUM VITAE

191

Curriculum Vitae

Angaben zur Person

Name Julia Maria Claudia Sterr

Dipl.-Ing. Univ. (Lebensmitteltechnologie)

Geburtsdatum 19. März 1982

Geburtsort Hutthurm, Deutschland

Akademische Ausbildung und beruflicher Werdegang

03/2017 – heute Leitung physikalisches Labor, Forschung & Entwicklung im

Bereich Extrusion bei Windmöller & Hölscher in Lengerich

08/2009 – 07/2016 Wissenschaftliche Angestellte am Lehrstuhl für

Lebensmittelverpackungstechnik an der Technischen

Universität München in Freising / Weihenstephan

08/2009 – heute Promotion am Lehrstuhl für Lebensmittelverpackungstechnik

an der Technischen Universität München in Freising /

Weihenstephan

Thema: “Influence of gases on the integrity of packaging

materials upon high pressure treatment”

10/2003 – 07/2009 Diplomstudium der Technologie und Biotechnologie der

Lebensmittel an der Technischen Universität München,

Freising/ Weihenstephan

Schwerpunkt: Verpackungstechnik

Abschluss: Diplom-Ingenieur Dipl.-Ing.

10/2002 – 08/2003 Studium Elektrotechnik an der Fachhochschule Deggendorf

Ohne Abschluss

08/1997 – 07/2002 Naturwissenschaftliches Gymnasium Adalbert-Stifter, Passau

Erwerb der allgemeinen Hochschulreife

192

Documents

Influence of gases on the integrity of packaging materials ...mediatum.ub.tum.de/doc/1353365/1353365.pdf · of the persons indicated they would spend $0.25 - $0.50 extra for an HPP